Intel’s return to boxed workstation processors arrived on February 2, 2026, when the company announced the Xeon 600 series, codenamed Granite Rapids-WS, with retail parts reaching shelves in late March. The launch folded the old Xeon W-2500 and W-3500 lines into a single family of 11 SKUs on the new W890 chipset and LGA4710 socket, topped by the 86-core Xeon 698X at $7,699. We have already spent time near the top of this family, reviewing the 64-core Xeon 696X inside HP’s Z8 Fury G6i; the Intel Xeon 658X is the lower-core counterpart, a 24-core, 48-thread part that currently sells for roughly $2,000 to $2,300 at retail, above its $1,869 list, and tested here on a bare lab platform rather than a tuned OEM workstation.

The 658X carries a 3.0 GHz base clock, 4.9 GHz max turbo, 144 MB of cache, and a 250W base power rating that climbs to 300W at maximum turbo. It pairs eight DDR5-6400 channels, a 4TB memory ceiling, and 128 PCIe 5.0 lanes with the AMX and AVX-512 acceleration that the whole family shares. Those platform figures are the same ones the 64- and 86-core parts carry; what changes as you move down the stack is core count, not memory reach or I/O. That makes the 658X a useful test of how much of the Xeon 600’s value lies in the platform rather than in the core count.

One piece of context is worth raising up front, with a caveat. A persistent round of enthusiast reporting and roadmap leaks holds that Granite Rapids-WS is a terminal platform: Intel has dropped mainstream Diamond Rapids from its Xeon roadmap, leaving Coral Rapids around 2028 as the next likely refresh. Intel has not confirmed any of this to us, and nothing official extends the workstation line beyond Granite Rapids-WS on the W890 socket. Buyers who care about a clean in-socket upgrade path should track how that shakes out, but for anyone evaluating the 658X on its own merits today, it changes little.

For this review, the Xeon 658X is installed in an Asus Pro WS W890E-SAGE SE board, giving us a look at the CPU across rendering, AI, compression, storage, and general system benchmarks.

Intel Xeon 600 Family Overview

SKU	Cores / Threads	Base / Boost	All-Core Turbo	L3 Cache	Base TDP	Mem Channels	Peak Memory	PCIe 5.0 Lanes	Boxed
Xeon 698X	86 / 172	2.0 / 4.8 GHz	3.0 GHz	336 MB	350 W	8	DDR5-8000 (MRDIMM)	128	—
Xeon 696X	64 / 128	2.4 / 4.8 GHz	3.5 GHz	336 MB	350 W	8	DDR5-8000 (MRDIMM)	128	Yes
Xeon 678X	48 / 96	2.4 / 4.9 GHz	3.8 GHz	192 MB	300 W	8	DDR5-8000 (MRDIMM)	128	Yes
Xeon 676X	32 / 64	2.8 / 4.9 GHz	4.3 GHz	144 MB	275 W	8	DDR5-8000 (MRDIMM)	128	Yes
Xeon 674X	28 / 56	3.0 / 4.9 GHz	4.3 GHz	144 MB	270 W	8	DDR5-8000 (MRDIMM)	128	—
Xeon 658X (review unit)	24 / 48	3.0 / 4.9 GHz	4.3 GHz	144 MB	250 W	8	DDR5-6400	128	Yes
Xeon 656	20 / 40	2.9 / 4.8 GHz	4.5 GHz	72 MB	210 W	8	DDR5-6400	128	—
Xeon 654	18 / 36	3.1 / 4.8 GHz	4.5 GHz	72 MB	200 W	8	DDR5-6400	128	Yes
Xeon 638	16 / 32	3.2 / 4.8 GHz	4.5 GHz	72 MB	180 W	4	DDR5-6400	80	—
Xeon 636	12 / 24	3.5 / 4.7 GHz	4.5 GHz	48 MB	170 W	4	DDR5-6400	80	—
Xeon 634	12 / 24	2.7 / 4.6 GHz	3.9 GHz	48 MB	150 W	4	DDR5-6400	80	—

All SKUs support DDR5-6400 (1 DPC) and up to 4TB of memory, CXL 2.0, and per-core AMX with FP16. X-series parts are unlocked. MRDIMM (DDR5-8000) is limited to the top five SKUs (674X and up). The 638, 636, and 634 drop to four memory channels and 80 PCIe lanes. Source: Intel.

Intel Xeon 658X Performance

Performance Testing

Review Unit Specifications

Our Intel Xeon 658X test platform consists of the following specifications:

• Motherboard: Asus Pro WS W890E-SAGE SE
• CPU: Intel Xeon 658X
• Memory: 128 GB DDR5-6400 ECC (4×32 GB)
• Storage: Samsung 9100 Pro 1TB
• Cooler: Noctua NH-U12S DX-4677

For context, we lined the Xeon 658X up against two higher-core workstation platforms we have recently tested. The HP Z8 Fury G6i represents the top of this same Xeon 600 family, running the 64-core Xeon 696X with 128GB of memory and two NVIDIA RTX PRO 6000 Max-Q GPUs. The Dell Precision 7875 covers the AMD side, built on the 96-core Threadripper PRO 9995WX with 512GB of memory and two RTX PRO 6000 cards. Both carry far more cores than the 24-core 658X, so these results are less about matching peak multi-threaded throughput and more about showing where a lower-core Xeon 600 part lands against the parts above and across from it. Since the focus here is on the CPU, GPU differences across platforms matter little for the benchmarks that follow.

Procyon AI Computer Vision

The Procyon AI Computer Vision Benchmark measures AI inference performance across CPUs, GPUs, and dedicated accelerators using a range of state-of-the-art neural networks. It evaluates tasks such as image classification, object detection, segmentation, and super-resolution using models that include MobileNet V3, Inception V4, YOLO V3, DeepLab V3, Real ESRGAN, and ResNet 50. Tests are run on multiple inference engines, including NVIDIA TensorRT, Intel OpenVINO, Qualcomm SNPE, Microsoft Windows ML, and Apple Core ML, providing a broad view of hardware and software efficiency. Results are reported for float- and integer-optimized models, providing a consistent, practical measure of machine vision performance for professional workloads.

CPU AI Computer Vision Overall Score

In the Procyon AI Computer Vision CPU benchmark, the Intel Xeon 658x platform achieved an overall score of 248, placing it ahead of both the HP Z8 Fury G6i (207) and the Dell Precision 7875 (157). The Xeon platform was about 20% faster than the HP system and roughly 58% ahead of the AMD-based Precision workstation, giving it the strongest CPU inference result in this group.

CPU Results	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
CPU Results
AI Computer Vision Overall Score	207	248	157
MobileNet V3	3.74 ms	1.26 ms	5.74 ms
ResNet 50	5.34 ms	4.58 ms	6.52 ms
Inception V4	17.16 ms	14.86 ms	20.42 ms
DeepLab V3	29.18 ms	26.29 ms	47.75 ms
YOLO V3	23.23 ms	26.20 ms	21.97 ms
REAL-ESRGAN	837.83 ms	1113.34 ms	1288.54 ms

Blender 4.5 CPU

Blender is an open-source 3D modeling application. This benchmark was run using the Blender Benchmark utility across CPU and GPU. The score is measured in samples per minute, with higher values indicating better performance.

In the Blender 4.5 CPU benchmark, the Intel Xeon 658x platform trailed both comparison systems across all three scenes. It posted 365.315 samples per minute in Monster, 234.081 in Junkshop, and 186.163 in Classroom. The HP Z8 Fury G6i was about 87% to 92% faster, depending on the scene, while the Dell Precision 7875 was much further ahead, with gains ranging from roughly 184% to 218%.

Blender CPU (samples per minute; higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
Monster	694.751	365.315	1039.121
Junkshop	449.356	234.081	744.601
Classroom	347.940	186.163	574.705

Blender 5.0 CPU

In the Blender 5.0 CPU benchmark, the Intel Xeon 658x platform again trailed the HP Z8 Fury G6i across all three scenes. The Xeon system reached 313.386 samples per minute in Monster, 241.790 in Junkshop, and 183.265 in Classroom. The HP system was about 90% to 93% faster across the workload set.

Blender CPU (samples per minute; higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)
Monster	606.140	313.386
Junkshop	467.820	241.790
Classroom	347.874	183.265

PCMark 10

PCMark 10 is an industry-standard benchmark that measures overall system performance in modern office environments. It features updated workloads for Windows 10 or 11 and evaluates everyday tasks such as productivity, web browsing, video conferencing, and content creation. The benchmark is easy to run, delivers multi-level scoring (from high-level overall scores to detailed workload scores), and includes dedicated battery-life and storage tests. While UL Solutions now recommends Procyon for newer application-based testing, PCMark 10 remains a reliable and widely used tool for assessing overall PC performance.

In PCMark 10, the Intel Xeon 658x platform posted an overall score of 9,657, placing it ahead of the HP Z8 Fury G6i (7,742) but behind the Dell Precision 7875 (11,433). The Xeon system was about 25% faster than the HP workstation, while the Dell system finished roughly 18% ahead.

PCMark10 (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
Overall Score	7,742	9,657	11,433

Blackmagic RAW Speed Test

The Blackmagic RAW Speed Test is a performance benchmarking tool that measures a system’s ability to handle video playback and editing with the Blackmagic RAW codec. It evaluates how well a system can decode and play back high-resolution video files, providing frame rates for both CPU- and GPU-based processing.

In the Blackmagic RAW Speed Test, the Intel Xeon 658x platform delivered 205 FPS in the 8K CPU test and 181 FPS in the 8K GPU test. Its CPU result was about 34% behind the HP Z8 Fury G6i but roughly 30% ahead of the Dell Precision 7875. The GPU result was weaker, trailing the HP by about 72% and the Dell by roughly 52%.

Blackmagic RAW (Higher FPS is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
8K CPU	311	205	158
8k GPU	650	181	276

3DMark CPU

The 3DMark CPU Profile evaluates processor performance across six threading levels: 1, 2, 4, 8, 16, and max threads. Each test runs the same boid-based simulation workload to assess how well the CPU scales under different thread counts, with minimal GPU involvement. The benchmark helps identify single-threaded efficiency and multithreaded potential for tasks such as gaming, content creation, and rendering. Scores on 8 threads often align with modern DirectX 12 gaming performance, while 1–4-thread results reflect older or esports scenarios.

The 3DMark CPU Profile benchmark showed the Intel Xeon 658x platform ahead of the HP Z8 Fury G6i across every thread count, though the Dell Precision 7875 took the lead in the heavier multi-threaded tests. At Max Threads, the Xeon scored 16,890, about 7% ahead of the HP and roughly 39% behind the Dell. The Xeon also kept a smaller advantage over the HP in the lower-thread tests, including 990 at 1 thread and 13,022 at 16 threads.

3DMark CPU (Higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
Max Threads	15,792	16,890	27,670
16 Threads	11,241	13,022	15,378
8 Threads	6,635	7,185	8,477
4 Threads	3,594	3,751	4,701
2 Threads	1,816	1,944	2,378
1 Threads	895	990	1,237

Geekbench 6

Geekbench 6 is a cross-platform benchmark that measures overall system performance.

In Geekbench 6, the Intel Xeon 658x platform delivered CPU results very close to those of the HP Z8 Fury G6i, while trailing the AMD-powered Dell Precision 7875. The Xeon scored 2,383 in single-core and 21,447 in multi-core performance, putting it about 2% ahead of the HP in single-core and roughly 1.6% ahead in multi-core. The Dell Precision 7875 was still well ahead, with leads of about 36% in single-core and 33% in multi-core performance.

GeekBench (Higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
CPU Single Core	2,333	2,383	3,240
CPU Multi-Core	21,110	21,447	28,618

Y-Cruncher

y-cruncher is a multithreaded and scalable program that can compute Pi and other mathematical constants to trillions of digits. Since its launch in 2009, it has become a popular benchmarking and stress-testing application for overclockers and hardware enthusiasts.

In Y-Cruncher, the Intel Xeon 658x platform had a mixed showing, starting behind the HP Z8 Fury G6i but still beating the Dell Precision 7875 in the smaller 250-million- and 500-million-digit runs. The Xeon completed 250 million digits in 1.711 seconds and 500 million digits in 3.758 seconds, faster than the Dell but around 41% to 42% behind the HP. As the workload increased, the Xeon fell behind both comparison systems, with the 5-billion-, 10-billion-, and 25-billion-digit tests taking 54.207 seconds, 118.898 seconds, and 326.454 seconds, respectively.

Y-Cruncher (lower duration is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
250 Million	1.203 s	1.711 s	2.369 s
500 Million	2.667 s	3.758 s	4.281 s
1 Billion	6.042 s	8.312 s	7.617 s
2.5 Billion	17.047 s	23.065 s	15.188 s
5 Billion	37.890 s	54.207 s	29.795 s
10 Billion	82.983 s	118.898 s	61.572 s
25 Billion	232.832 s	326.454 s	169.289 s
50 Billion	N/A	N/A	371.039 s
100 Billion	N/A	N/A	844.503 s

Y-Cruncher BBP

The Y-Cruncher BBP test uses the Bailey-Borwein-Plouffe formula, which extracts binary digits of Pi at a specific position without computing the digits that come before it. Unlike the main swap-mode runs that lean heavily on the memory subsystem, BBP digit extraction is almost entirely compute-bound and highly parallel, so it scales with raw core throughput rather than memory bandwidth. Lower durations are better.

In the Y-Cruncher BBP runs, the 24-core Xeon 658X completed the 1 billion, 10 billion, and 100 billion digit tests in 0.506, 5.415, and 59.520 seconds. That is roughly half the throughput of the 64-core Xeon 696X in the HP Z8 Fury G6i (0.333, 2.806, and 29.897 seconds), in line with the core-count gap between the two. The Dell Precision 7875 did not get BBP testing.

Y-Cruncher BBP (lower duration is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)
1 Billion BBP	0.333 s	0.506 s
10 Billion BBP	2.806 s	5.415 s
100 Billion BBP	29.897 s	59.520 s

7-Zip Compression

The 7-Zip Compression Benchmark evaluates CPU performance during compression and decompression, measuring GIPS (Giga Instructions Per Second) and CPU usage. Higher GIPS and efficient CPU usage indicate superior performance.

In the 7-Zip Compression Benchmark, the Intel Xeon 658x platform delivered a Total Rating of 250.083 GIPS. That placed it behind the HP Z8 Fury G6i, which was about 38% faster overall, but far ahead of the Dell Precision 7875 in this specific result set. The Xeon also posted a Resulting Compression Rating of 233.557 GIPS and a Resulting Decompression Rating of 266.608 GIPS, again trailing the HP but finishing well ahead of the Dell numbers shown here.

7-Zip Compression Benchmark (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x 48C, 2.4GHz | 128GB RAM)	Intel Xeon 658X Test Platform (24C, 3.0GHz | 128GB RAM)	Dell Precision 7875 (AMD 9995WX 96C, 2.5GHz | 512GB RAM)
Compression
Current CPU Usage	5,483%	4,038%	6,445%
Current Rating/Usage	6.247 GIPS	5.786 GIPS	6.949 GIPS
Current Rating	342.522 GIPS	233.641 GIPS	48.392 GIPS
Resulting CPU Usage	5,461%	4,037%	701%
Resulting Rating/Usage	6.242 GIPS	5.705 GIPS	7.010 GIPS
Resulting Rating	340.875 GIPS	233.557 GIPS	49.108 GIPS
Decompression
Current CPU Usage	6,029%	4,688%	728%
Current Rating/Usage	5.839 GIPS	5.705 GIPS	6.801 GIPS
Current Rating	352.023 GIPS	267.475 GIPS	49.526 GIPS
Resulting CPU Usage	5,990%	4,657%	749%
Resulting Rating/Usage	5.827 GIPS	5.725 GIPS	6.832 GIPS
Resulting Rating	349.054 GIPS	266.608 GIPS	51.181 GIPS
Total Rating
Total CPU Usage	5,726%	4,347%	725%
Total Rating/Usage	6.034 GIPS	5.755 GIPS	6.921 GIPS
Total Rating	344.964 GIPS	250.083 GIPS	50.145 GIPS

Conclusion

The Intel Xeon 658X is a platform play, not a core-count play, and the numbers align with that. It is a 24-core part tested against the 64-core Xeon 696X in the HP Z8 Fury G6i and the 96-core AMD Threadripper PRO 9995WX in the Dell Precision 7875, so the heavily threaded losses are no surprise. In Blender 4.5 CPU, it trailed both, with the HP roughly 87% to 92% faster and the Dell 184% to 218% ahead across the three scenes. 7-Zip ran in the same direction; its total of 250.083 GIPS is about 38% behind the HP. When a workload scales across two to four times the number of cores, a 24-core chip loses, and this one does too.

What it does not lose is the work that relies on the platform and its cores. The 658X led the entire group in Procyon AI Computer Vision CPU at 248, about 20% ahead of the HP and 58% ahead of the Dell, a result that tracks with the FP16-capable AMX units in every Xeon 600 core. It also edged the 64-core HP in general-system and lightly-threaded tests despite the core deficit: 9,657 in PCMark 10 (about 25% over the HP, though 18% behind the Dell), a lead at every 3DMark CPU Profile thread count, and a near tie with the same-generation 696X in Geekbench 6 at 2,383 single-core and 21,447 multi-core. Single-thread parity with its 64-core sibling is the tell, since both run the same Redwood Cove cores.

The hardware behind those results is consistent across the stack. The 658X carries 24 cores, 48 threads, 144 MB of cache, eight channels of DDR5-6400, a 4TB memory ceiling, and 128 PCIe 5.0 lanes, the same I/O and memory reach Intel gives the 64- and 86-core parts above it. It does not open a memory-channel lead over the Threadripper PRO platform, since both run eight channels at DDR5-6400. What Intel sells here is capacity and lane count for the money, not bandwidth that the competition cannot match.

One caveat belongs in the buying decision, with a qualifier. Granite Rapids-WS looks like a limited-life platform: enthusiast reports indicate that Intel has dropped mainstream Diamond Rapids, with no workstation successor in sight, leaving Coral Rapids around 2028 as the next likely refresh, and Intel has not confirmed any of that. For a self-built rig bought as a future upgrade path on W890, that uncertainty matters. For the way most of these chips will actually ship, inside a complete HP, Dell, or Lenovo workstation, it matters far less, since those buyers replace whole systems on a refresh cycle rather than dropping a new CPU into an old socket. In that context, the 658X is a sound choice. It is not the part for heavy multi-threaded rendering, but it is a lower-core entry into Intel’s new workstation platform with full I/O and memory, group-leading CPU AI inference, and solid general-system results for builders who value memory capacity and PCIe lanes over raw cores.

Product Page – Intel Xeon 658X

The post Intel Xeon 658X Review: 24 Cores Into Intel’s New Workstation Platform appeared first on StorageReview.com.

AMD continues to strengthen its position in the global high-performance computing (HPC) market, with the latest TOP500 and Green500 rankings showing broad adoption of AMD EPYC processors and Instinct accelerators across leading supercomputing deployments.

According to the June 2026 TOP500 list, AMD technology now powers 191 systems, an 11% year-over-year increase, and accounted for 41% of the new systems added to this edition. The company also claims four of the world’s 10 fastest supercomputers and four of the 10 most energy-efficient systems, underscoring the growing role of AMD platforms in both AI and traditional HPC environments.

Strong Showing in TOP500 Rankings

AMD-powered systems remain well represented among the world’s highest-performing supercomputers, even as the No. 1 spot changed hands this cycle. China’s LineShine, a 2.198 exaflop system at the National Supercomputing Centre in Shenzhen, debuted at the top of the June 2026 list and ended El Capitan’s run as the world’s fastest supercomputer. AMD-powered machines still hold three of the top six: El Capitan at No. 2, Frontier at No. 3, and Eni’s newly deployed HPC7 at No. 6.

These deployments rely on combinations of AMD EPYC CPUs and AMD Instinct GPUs to support large-scale scientific computing, simulation, and AI workloads. The continued presence of AMD-powered systems near the top of the TOP500 rankings reflects the industry’s demand for architectures capable of supporting increasingly complex data-intensive applications.

Energy Efficiency Remains a Key Differentiator

The Green500 rankings, which measure supercomputer energy efficiency, have become an increasingly important benchmark as power consumption emerges as a primary constraint for future HPC and AI deployments.

AMD-powered systems secured four positions in the Green500 top 10, including Otus (No. 5), Capella (No. 6), AMD Ouranos (No. 9), and Portage (No. 10). AMD technology also powers more than half of the Green500 top 50 systems, accounting for 56% of the list.

The results highlight the industry’s growing focus on balancing computational performance with operational efficiency, particularly as AI training and inference workloads continue to scale.

Growing Role in European HPC and Sovereign AI Initiatives

AMD is also expanding its footprint across Europe as governments, research institutions, and enterprises invest in sovereign AI and next-generation HPC infrastructure.

Eni’s HPC7 supercomputer, ranked sixth globally, represents one of Europe’s largest industrial HPC deployments. The system builds on the company’s earlier HPC6 platform and supports AI, simulation, and energy research workloads.

The University of Cambridge recently announced two systems powered by AMD Instinct MI355X GPUs, marking the first TOP500 deployments based on the accelerator. The systems entered the rankings at positions 67 and 68.

LUMI, hosted by CSC in Finland and operated through the EuroHPC Joint Undertaking, remains one of Europe’s flagship HPC resources. Ranked No. 11 globally, the system supports a broad range of scientific computing and AI research projects across the region.

In France, GENCI is advancing plans for Alice Recoque, the country’s first exascale supercomputer. The system will combine AMD Instinct MI430X GPUs with 6th Generation AMD EPYC processors and is expected to serve both traditional HPC and AI workloads as part of France’s broader AI infrastructure strategy.

AMD Previews MI430X Accelerator

At the HPC User Forum 2026, AMD provided an early look at its upcoming Instinct MI430X GPU, targeting organizations that require both AI acceleration and high-precision scientific computing.

While AI workloads often emphasize lower-precision formats, many scientific applications continue to depend on double-precision (FP64) performance. Climate modeling, computational fluid dynamics, advanced materials research, aerospace design, and fusion energy simulations all require high levels of numerical accuracy.

AMD projects the Instinct MI430X will deliver more than 200 TFLOPS of native FP64 performance, positioning it as a potential flagship accelerator for next-generation leadership-class HPC systems. If achieved, that level of performance would establish a new benchmark for large-scale simulation, modeling, and AI-assisted scientific computing.

The latest TOP500 and Green500 results suggest that AMD’s strategy of pairing high-core-count EPYC processors with increasingly capable Instinct accelerators continues to gain traction across both traditional supercomputing and emerging AI infrastructure deployments.

The post AMD Powers 4 of the Top 10 on the June 2026 TOP500 as China’s LineShine Takes No. 1 appeared first on StorageReview.com.

At ISC 2026, DDN announced a broad expansion of its AI and HPC data platform portfolio, introducing new hardware, software, and platform capabilities designed to improve GPU utilization, accelerate AI inference, and simplify large-scale AI infrastructure deployments.

The announcements include the AI400X3M storage appliance, the general availability of DDN’s distributed KV Cache acceleration technology integrated with NVIDIA Dynamo, and new security, observability, and infrastructure management capabilities across the company’s AI data platforms.

DDN said the updates are intended to address bottlenecks throughout the AI data pipeline, including data preparation, model training, inference, retrieval-augmented generation (RAG), reasoning models, and agentic AI workloads. The company is focusing on improving infrastructure efficiency by increasing GPU utilization while lowering power consumption and infrastructure costs.

DDN CEO and Co-Founder Alex Bouzari said modern AI infrastructure requires more than compute resources alone. He emphasized that efficient data management, security, and operational visibility have become critical components of production AI deployments, with the company’s latest releases aimed at improving GPU efficiency, reducing inference costs, and accelerating time-to-first-token.

AI400X3M Targets AI and HPC Performance Density

Leading the announcement is the AI400X3M, the latest appliance built on DDN’s EXAScaler parallel file system.

AI400X3M Specifications

Performance
Sequential Read Performance	Up to 190 GB/s
Sequential Write Performance	Up to 110 GB/s
Random Read IOPS	8M
System Features
Platform	Turnkey EXAScaler shared parallel file system appliance for AI with Active/Active storage controllers
Host Ports Per Appliance	4x XDR/400GbE OSFP or 8x NDR200/200GbE QSFP112
Drive Support	24x 2.5″ dual port, hot swappable PCIe G5 NVMe
Capacity Options	120/250/500TB and 1PB/2PB usable
Standard Firmware Features	LUN mapping and masking, intelligent write striping, read QoS, data integrity check/correction, interface options (CLI, GUI, Python API), outbound syslog including RELP support, state change messages (via email and SNMP traps)
Safety & Compliance
Safety	IEC/EN/UL/CSA 62368-1, GB4943
EMC	EN 55032 Class A, EN 55035, EN 61000-3-2, EN 61000-3-3, FCC Part 15 Class A, VCCI Class A, ICES-003 Class A, GB/T9254 Class A, BSMI Class A
Environmental	RoHS, REACH, WEEE, ERP Lot9X
Physical Attributes
Form Factor	2U rack mount
Dimensions	Height: 3.43″ (87mm) Width: 17.56″ (446mm) Depth: 33.7″ (856mm) without bezel; 35.2″ (895mm) overall Adjustable rack rail: 26.75″ to 34.0″ (680–860mm)
Weight	98 lbs (44kg) empty; 108 lbs (49kg) max
Power & Cooling
Input Voltage	200–240V 50/60 Hz
Power Supply	2x hot swappable, redundant, IEC60320-C20 inlet
Operating Environment
Temperature Range	5°C to 30°C (41°F to 86°F)
Relative Humidity	8%–80% non-condensing
Altitude	3,117 ft (950m) @ 30°C — 10,000 ft (3,048m) @ 23°C

 

The platform delivers up to 190GB/s of throughput and up to 35% higher read performance than the previous generation. DDN also supports configurations that scale to 30PB within a single rack, while adding hybrid disk support that combines flash and HDD storage to balance performance, capacity, and cost.

The appliance is designed for AI model training, inference, checkpointing, HPC simulation, and other highly parallel workloads where storage throughput can directly impact GPU utilization. General availability is expected by the end of the third quarter of 2026.

Distributed KV Cache Now Generally Available

Following an earlier technology preview at NVIDIA GTC 2026, DDN officially launched its distributed KV Cache acceleration architecture, integrating NVIDIA Dynamo. The capability is available across both the Infinia object storage platform and the EXAScaler file system.

Rather than relying on local GPU memory for model context, the distributed architecture stores and retrieves KV cache data from DDN’s data platform, reducing memory bottlenecks during inference.

DDN said the platform supports NVIDIA Dynamo, vLLM, and other modern inference frameworks while enabling shared distributed KV cache across multiple inference nodes. The company claims up to 55 times faster KV cache loading for large-scale inference workloads, along with improved GPU utilization, faster token generation, and lower infrastructure costs for large language model deployments.

The technology is intended to benefit long-context inference, retrieval-augmented generation, reasoning models, and agentic AI workflows that repeatedly access large model contexts.

Unified AI Data Pipeline

DDN also outlined its broader strategy of combining Infinia object storage for inference with EXAScaler for model training and checkpointing under a single AI data infrastructure.

The company said Infinia provides low-latency metadata operations, high concurrency, and object performance optimized for inference-heavy workloads, while EXAScaler continues to serve high-performance training environments. Together, the platforms are intended to eliminate data silos and maintain high GPU utilization throughout the AI lifecycle.

New Security and Infrastructure Features

Additional platform enhancements focus on enterprise deployment requirements, including bare-metal multi-tenancy, KMIP-based encryption and key management, VictoriaLogs integration for operational visibility, and expanded multi-tenant APIs.

On the storage side, DDN introduced intelligent file pinning and NAND-accelerated Hot Pools, allowing frequently accessed data to remain on flash while automatically tiering colder data to lower-cost hard drives. The company said these features are designed to improve storage efficiency while reducing flash capacity requirements.

Continued Cloud AI Expansion

DDN also highlighted continued adoption of its cloud AI infrastructure offerings, including Managed Lustre developments announced with Google Cloud and a Salesforce deployment that uses Google Cloud Managed Luster powered by EXAScaler.

According to DDN, the Salesforce implementation achieved 1.5 times faster model training, reduced I/O latency by 75%, and lowered training costs by 42%, demonstrating the impact of reducing storage bottlenecks in large-scale AI environments.

With the latest announcements, DDN continues to expand its portfolio beyond parallel storage into AI data orchestration, inference acceleration, and infrastructure management, positioning its platforms to support enterprise, cloud, research, and sovereign AI deployments operating at large GPU scale.

The post DDN Launches AI400X3M and Dynamo-Integrated KV Cache Acceleration at ISC 2026 appeared first on StorageReview.com.

Micron has announced a strategic agreement with Anthropic focused on AI memory and storage architecture, long-term component supply, enterprise deployment of Anthropic’s Claude models within Micron, and participation in Anthropic’s Series H funding round.

The partnership aligns infrastructure development more closely with the requirements of large-scale AI training and inference environments. The companies plan to collaborate on memory and storage subsystem design and optimization, while also establishing a supply framework intended to support Anthropic’s future compute expansion.

Micron Executive Vice President and Chief Business Officer Sumit Sadana said memory and storage technologies remain foundational to AI deployments spanning data centers and edge environments. He noted that combining Micron’s infrastructure expertise with Anthropic’s AI development efforts is intended to help advance next-generation AI platforms.

Anthropic co-founder and Chief Compute Officer Tom Brown highlighted memory and storage as critical components of the company’s compute strategy for training and serving Claude models. He said the collaboration is designed to optimize infrastructure for Anthropic workloads while providing a secure supply foundation as compute demand continues to grow.

Joint Focus on Memory and Storage Optimization

At the center of the agreement is a technical collaboration focused on improving memory and storage performance for frontier AI workloads. The companies will evaluate how memory and storage subsystems perform across a range of training and inference scenarios and examine interactions throughout the broader infrastructure stack.

Micron’s portfolio of high-bandwidth memory (HBM), DRAM, and data center SSDs will serve as the foundation for this work, underpinning performance, power efficiency, and total cost of ownership across AI training and inference. The companies expect the effort to drive advances in memory and storage performance, energy efficiency, and the token economics of Anthropic’s AI infrastructure.

As AI models continue to scale, memory bandwidth, capacity, storage throughput, and energy efficiency have become increasingly important constraints. The collaboration is intended to identify opportunities for architectural improvements that better align infrastructure resources with AI workload requirements.

Multi-Year Supply Agreement

Alongside the engineering collaboration, the companies have entered into a memory and storage supply agreement covering Micron’s data center product portfolio.

The agreement is designed to support Anthropic’s long-term compute roadmap as the company expands training and inference capacity for future generations of Claude models. Securing access to memory and storage components has become a strategic consideration for AI developers as demand for infrastructure continues to grow across the industry.

Claude Deployment Across Micron

Micron also disclosed that it has deployed Anthropic’s Claude models internally across engineering, manufacturing, and enterprise operations.

The company said Claude is being used to accelerate software development and support more advanced agentic workflows. Micron reports that these deployments have contributed to productivity improvements across complex business and technical functions and expects broader AI adoption to influence future design, development, and operational processes.

Strategic Investment

As part of the broader relationship, Micron made a strategic investment in Anthropic’s Series H round, the $65 billion raise announced in May 2026 that valued Anthropic at $965 billion post-money. Micron joined fellow memory and chip suppliers Samsung and SK hynix as strategic infrastructure partners in the round, complementing the technical and supply agreements and reflecting a shared focus on the infrastructure required to support increasingly demanding AI workloads.

The post Micron and Anthropic Form Strategic AI Infrastructure Partnership appeared first on StorageReview.com.

For most of the last decade, the high-core workstation conversation has been largely led by AMD. Threadripper PRO pushed core counts, cache, and PCIe lanes past what Intel’s Xeon-W line could offer; the prior Xeon-W flagship topped out at 60 cores, while AMD kept climbing. Intel’s Xeon 600 series, launched in February of this year, is the first CPU in years built to take that argument back, reaching 86 cores on the Granite Rapids-WS platform with 128 PCIe 5.0 lanes. The HP Z8 Fury G6i is the system HP built around it.

HP frames the Z8 Fury G6i as an AI workstation rather than a CAD tower, which aligns with current industry trends. One Xeon 600 processor pairs with up to four NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition GPUs (300-watt) for 384 GB of aggregate VRAM, or a single 600W RTX PRO 6000 Workstation Edition for peak single-card performance. The platform supports up to 2 TB of DDR5-6400 ECC memory across eight channels, four M.2 drive slots (with options for more), and up to nine PCIe slots in a roughly 54-liter tower. HP also offers a rack mounting kit (5RU), so a team can centralize it, should the use case dictate.

Our review unit sits one rung below the top of the stack. It runs the 64-core Xeon 696X with two RTX PRO 6000 Max-Q cards for 192 GB of combined VRAM, 128 GB of DDR5-6400, and a four-drive NVMe layout that separates a fast Gen5 boot volume from three Gen4 data drives. That configuration frames the question this review works through: how a single high-core Xeon and a dense Blackwell GPU config balance against each other across professional graphics, GPU rendering, and AI inference, and where the platform gives ground.

The Z8 Fury G6i is configurable on the HP website and starts at roughly $7900 at the time of this review. Our configuration comes in at $74,878. It’s worth noting that most of these systems are bought through corporate acquisition, and volume pricing will be better.

Specifications

Specification	HP Z8 Fury G6i
Processor Options (Intel W890 Chipset)
Flagship Model	Intel Xeon 698X: 86 cores, 172 threads, 2.0GHz base, up to 4.8GHz Turbo Boost, 336MB L3 cache, 350W
High-Core Options	Intel Xeon 696X: 64 cores, 128 threads, 2.4GHz base, up to 4.8GHz Turbo, 336MB L3 cache, 350W Intel Xeon 678X: 48 cores, 96 threads, 2.4GHz base, up to 4.9GHz Turbo, 192MB L3 cache, 300W
Performance Options	Intel Xeon 676X: 32 cores, 64 threads, 2.8GHz base, up to 4.9GHz Turbo, 144MB L3 cache, 275W Intel Xeon 674X: 28 cores, 56 threads, 3.0GHz base, up to 4.9GHz Turbo, 144MB L3 cache, 270W Intel Xeon 658X: 24 cores, 48 threads, 3.0GHz base, up to 4.9GHz Turbo, 144MB L3 cache, 250W Intel Xeon 656: 20 cores, 40 threads, 2.9GHz base, up to 4.8GHz Turbo, 72MB L3 cache, 210W Intel Xeon 654: 18 cores, 36 threads, 3.1GHz base, up to 4.8GHz Turbo, 72MB L3 cache, 200W
Memory & Storage
System Memory	16 DIMM slots; Up to 2TB DDR5-6400 ECC Registered Memory
Total Storage Capacity	Up to 104TB total storage
Internal NVMe Slots	Supports up to eight PCIe M.2 SSD devices
Front Accessible Storage	Up to four front-accessible hot-swappable NVMe drives with LED indicators and email notifications
SATA Support	4TB-12TB 7200RPM SATA Enterprise HDD support; optional slim DVD-ROM/DVD-Writer
Available Graphics
Ultra High-End	NVIDIA A800 (40GB GDDR6) NVIDIA RTX PRO 6000 Blackwell Generation (96GB GDDR7)
High-End	NVIDIA RTX PRO 5000 Blackwell Generation (48GB GDDR7) NVIDIA RTX PRO 4500 Blackwell Generation (32GB GDDR7)
Mid-Range	NVIDIA RTX PRO 4000 Blackwell Generation (24GB GDDR7) NVIDIA RTX PRO 2000 Blackwell Generation (16GB GDDR7)
Entry	NVIDIA RTX A1000 (8GB GDDR6)
I/O & Networking
Front Ports	4x USB Type-A 5Gbps (1 charging) Optional premium front I/O with 2x USB-C 20Gbps 1x headphone/microphone combo jack
Rear Ports	1x USB Type-C 10Gbps 5x USB Type-A 5Gbps Optional dual Thunderbolt 5 USB-C 40Gbps ports
Networking	Integrated Intel I219-LM PCIe GbE Optional 10GbE / 25GbE networking modules and NICs Optional Wi-Fi 7 and Bluetooth 5.4
Certifications & Software
ISV Certifications	Certified for professional applications and advanced workstation workflows
HP Software Suite	HP Anyware Pro HP Z Remote Graphics Software (RGS) HP Support Assistant HP Smart Sense
Security & Management	HP Wolf Security HP Sure Start HP Sure Click HP Sure Sense TPM 2.0 HP BIOSphere
Sustainability & Efficiency	EPEAT Gold certified ENERGY STAR configurations available 60% recycled plastics 20% recycled steel 80 Plus Platinum power supplies
Physical Specifications
Dimensions (H x W x D)	17.5 x 8.6 x 22 in (44.5 x 21.95 x 55.9 cm) up to 17.5 x 10 x 22 in (44.5 x 25.35 x 55.9 cm) with max side panel
Weight	Starting at 48.9 lb (22.2 kg)
Power Supply	1350W, 1700W, or 2700W PSU options Redundant and aggregate power configurations are available

Design and Build

The Z8 Fury G6i carries over the look HP has settled on across its latest workstation lineup, with a uniform matte black finish from the chassis to the front fascia. The face is dominated by a plastic diamond-mesh grille that runs the full height of the tower for airflow, broken only by the front I/O strip near the top and the metallic HP logo lower down. The tower itself is substantial: it starts at 48.9 lb and measures 17.5 x 8.6 x 22 inches without the rear handle, growing to 17.5 x 10 x 22 inches with the maximum side panel fitted. That heft is a function of the dual-PSU, multi-GPU support built inside, but the result is a rigid, well-damped enclosure that feels every bit the professional-grade workstation it is.

Storage

For boot and high-speed flash storage, the Z8 Fury G6i provides four onboard PCIe Gen5 M.2 slots (labeled SSD0 through SSD3), each fitted with a finned heatsink and a tool-free blue latch for retention. HP sells the drives in 1, 2, 4, and 8TB capacities, so the four slots can be populated to suit anything from a single boot drive to a high-capacity NVMe array.

For bulk storage, the Z8 Fury G6i includes two internal 3.5-inch drive bays with tool-free carriers, letting you slot in high-capacity HDDs without a screwdriver. The bays sit on a backplane with the SATA/power connectors fixed in place, so drives seat directly as they slide in.

I/O and Expansion

The Z8 Fury G6i has a fairly standard set of I/O ports for workstations, with four USB Type-A 5 Gbps ports (one charging) and a single 3.5 mm headphone/microphone combo port. HP, however, offers a premium front I/O setup with two USB Type-A 5 Gbps ports (one charging), two USB Type-C 20 Gbps ports (both charging), and a 3.5 mm headphone/microphone combo port. Both front I/O configurations also offer an SD card reader alongside the I/O ports. Our review unit included the premium front I/O package.

On the rear of the Z8, we again see a fairly standard I/O setup with an integrated GbE Ethernet port, five USB Type-A 5 Gbps ports, a single USB Type-C 10 Gbps port, and another headphone/microphone combination port. Near our standard I/O ports, we also see the Flex I/O port, which offers up to 10GBASE-T or 2 Thunderbolt 5 ports. Also on the rear is a collapsible handle that folds flush against the chassis when not in use, but pops out to provide a sturdy hold point for lifting or repositioning the workstation.

When it comes to expansion slots, the Z8 Fury G6i has a total of 9 PCIe slots: 4 PCIe 5 x16, 3 PCIe 5 x8, 1 PCIe 5 x4, and 1 PCIe 4 x4. These four PCIe 5 x16 slots are what allow the Z8 to house the quad NVIDIA RTX PRO 6000 Blackwell Max-Q cards in a fully loaded configuration.

The board also exposes a dedicated network MCIO connector that accepts an optional add-in module for 2x 10GbE or 2x 25GbE LANs. Our review unit shipped without the module, leaving the slot open, but it’s the path HP provides for adding higher-speed networking without consuming a standard PCIe expansion slot.

HP pairs a tool-free PCIe retention latch at the top slot with a pivoting bar that swings to release the card, so the top GPU disengages cleanly without fighting the PCIe slot lock.

We can also see to the left of the chassis the dual removable power supplies that can be configured in either a redundant or cumulative configuration, totaling up to 2700W, to feed configurations like the highest spec buildout that contains a 350W TDP CPU and up to 1200W of GPUs, being either a single RTX PRO 6000 Blackwell card or 4x RTX Pro 6000 Blackwell Max-Q (300W). The power supply setup is unique in that it can support high-end configurations that would exceed what a 15-A or 20-A 120V circuit can handle on its own, before requiring a move to a 240V circuit. For areas that could supply two discrete 120V circuits, you can run the hardware in that environment off this dual-PSU configuration

Performance Testing

Review Unit Specifications

Our HP Z8 Fury G6i review unit arrived at the lab with the following specifications:

• CPU: Intel Xeon 696X (64c/128t)
• GPU: 2x NVIDIA RTX PRO 6000 Blackwell Max-Q Workstation Edition
• RAM: 128 GB DDR5-6400 ECC (4×32 GB)
• Storage:
  • Boot Drive: 1x 2 TB HP Z Turbo Drive PCIe 5×4 M.2 SSD
  • Data Drives: 3x 2 TB HP Z Turbo Drive PCIe 4×4 TLC M.2 SSD

Comparison Specifications

For comparative results, we have lined up the HP Z8 Fury G6i against our Intel Xeon 658x test platform, which is loaded with the same 128 GB of DDR5 RAM and an NVIDIA RTX 4090. This platform has a Xeon 6-series workstation CPU in the same class as the Z8, but with a lower core count of 24 cores/48 threads. We have also set the Z8 against our previously reviewed Dell Precision 7875, which features the 96-core/192-thread AMD Threadripper 9995WX, 512 GB of DDR5-5200 ECC RAM, and dual NVIDIA RTX PRO 6000 Blackwell GPUs.

Procyon AI Computer Vision

The Procyon AI Computer Vision Benchmark measures AI inference performance across CPUs, GPUs, and dedicated accelerators using a range of state-of-the-art neural networks. It evaluates tasks such as image classification, object detection, segmentation, and super-resolution using models that include MobileNet V3, Inception V4, YOLO V3, DeepLab V3, Real ESRGAN, and ResNet 50. Tests are run on multiple inference engines, including NVIDIA TensorRT, Intel OpenVINO, Qualcomm SNPE, Microsoft Windows ML, and Apple Core ML, providing a broad view of hardware and software efficiency. Results are reported for float- and integer-optimized models, providing a consistent, practical measure of machine vision performance for professional workloads.

CPU AI Computer Vision Overall Score

In the Procyon AI Computer Vision CPU benchmark, the HP Z8 Fury G6i achieved an overall score of 207, placing it between the Intel Xeon 658x platform (248) and the Dell Precision 7875 (157). The HP system trailed the Xeon platform by 16.5%, while outperforming the AMD-based Precision workstation by 31.8%, demonstrating strong CPU inference performance across a range of computer vision models.

GPU AI Computer Vision Overall Score

Using its dual NVIDIA RTX PRO 6000 Max-Q GPUs, the HP Z8 Fury G6i posted a Procyon AI Computer Vision GPU score of 1,151. While this was lower than the Dell Precision 7875’s 1,619 score, the HP platform still delivered substantial AI acceleration, finishing approximately 29% behind the Dell workstation in overall GPU-based computer vision performance.

CPU Results	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
CPU Results
AI Computer Vision Overall Score	207	248	157
MobileNet V3	3.74 ms	1.26 ms	5.74 ms
ResNet 50	5.34 ms	4.58 ms	6.52 ms
Inception V4	17.16 ms	14.86 ms	20.42 ms
DeepLab V3	29.18 ms	26.29 ms	47.75 ms
YOLO V3	23.23 ms	26.20 ms	21.97 ms
REAL-ESRGAN	837.83 ms	1113.34 ms	1288.54 ms
GPU Results
AI Computer Vision Overall Score	1,151	N/A	1,619
MobileNet V3	0.61 ms	N/A	0.45 ms
ResNet 50	0.96 ms	N/A	0.82 ms
Inception V4	2.31 ms	N/A	2.16 ms
DeepLab V3	21.00 ms	N/A	6.60 ms
YOLO V3	4.74 ms	N/A	3.48 ms
REAL-ESRGAN	49.81 ms	N/A	47.33 ms

Blender 4.5 CPU

Blender is an open-source 3D modeling application. This benchmark was run using the Blender Benchmark utility across CPU and GPU. The score is measured in samples per minute, with higher values indicating better performance.

In the Blender CPU benchmark, the HP Z8 Fury G6i delivered mixed but competitive results. Compared to the Intel Xeon 658x platform, the HP system was substantially faster, posting gains of approximately 90% across all three scenes. However, the AMD-powered Dell Precision 7875 remained the performance leader, outperforming the HP by roughly 50–66% depending on the workload. Even so, the Z8 Fury established itself as a strong CPU rendering platform, comfortably outperforming the Xeon comparison system while narrowing the gap to the high-core-count Threadripper Pro workstation.

Blender CPU (samples per minute; higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Monster	694.751	365.315	1039.121
Junkshop	449.356	234.081	744.601
Classroom	347.940	186.163	574.705

Blender 4.5 GPU

When rendering on the GPU, the HP Z8 Fury G6i and Dell Precision 7875 were effectively neck-and-neck. The HP system held a slight advantage in the Monster (+1.3%) and Junkshop (+1.2%) scenes, while the Dell workstation edged ahead by less than 0.5% in Classroom. Overall, GPU rendering performance between the two dual RTX PRO 6000 platforms was essentially identical, with differences small enough to fall within normal benchmark variance.

Blender GPU (samples per minute; higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Monster	7351.497	7259.413
Junkshop	3992.506	3943.343
Classroom	3648.637	3665.272

PCMark 10

PCMark 10 is an industry-standard benchmark that measures overall system performance in modern office environments. It features updated workloads for Windows 10 or 11 and evaluates everyday tasks such as productivity, web browsing, video conferencing, and content creation. The benchmark is easy to run, delivers multi-level scoring (from high-level overall scores to detailed workload scores), and includes dedicated battery-life and storage tests. While UL Solutions now recommends Procyon for newer application-based testing, PCMark 10 remains a reliable and widely used tool for assessing overall PC performance.

In PCMark 10, which measures overall system responsiveness across common productivity, content creation, and office workloads, the HP Z8 Fury G6i posted a score of 7,742. This placed it behind both comparison systems, trailing the Intel Xeon 658x platform (9,657) by approximately 20% and the AMD-based Dell Precision 7875 (11,433) by roughly 32%. While the Z8 Fury is clearly optimized for professional workstations and accelerated compute workloads, the PCMark 10 results show that competing platforms deliver stronger performance across a broader mix of desktop-oriented tasks.

PCMark10 (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Overall Score	7,742	9,657	11,433

Blackmagic RAW Speed Test

The Blackmagic RAW Speed Test is a performance benchmarking tool that measures a system’s ability to handle video playback and editing with the Blackmagic RAW codec. It evaluates how well a system can decode and play back high-resolution video files, providing frame rates for both CPU- and GPU-based processing.

The HP Z8 Fury G6i turned in an impressive showing in the Blackmagic RAW Speed Test, leading both comparison systems in CPU and GPU decoding performance. In the 8K CPU test, the HP reached 311 FPS, outperforming the Intel Xeon 658x platform (205 FPS) by approximately 52% and nearly doubling the performance of the Dell Precision 7875 (158 FPS). The gap widened even further in the 8K GPU test, where the HP delivered 650 FPS, compared to 181 FPS from the Xeon platform and 276 FPS from the Precision 7875. These results highlight the Z8 Fury’s exceptional capability for high-resolution Blackmagic RAW playback and editing workflows.

Blackmagic RAW (higher FPS is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
8K CPU	311	205	158
8k GPU	650	181	276

Blackmagic Disk Speed Test

The Blackmagic Disk Speed Test evaluates storage performance by measuring read and write speeds, providing insights into a system’s ability to handle data-intensive tasks, such as video editing and large file transfers.

Storage performance was another area where the HP Z8 Fury G6i remained highly competitive. Equipped with a 2TB PCIe Gen5 HP Z Turbo boot drive and three 2TB PCIe Gen4 HP Z Turbo SSDs for data storage, the system delivered 8,911.5 MB/s read and 8,166.5 MB/s write performance on its primary drive. Compared to the Dell Precision 7875, which posted 9,111.4 MB/s read and 9,292.0 MB/s write, the HP trailed by just 2.2% in read throughput and by approximately 12.1% in write throughput.

The HP system also included a secondary storage volume that achieved 4,136.8 MB/s read and 5,149.3 MB/s write, providing ample bandwidth for active project data, scratch disks, and large media workloads. While the Dell workstation held a modest advantage on the primary drive benchmark, the Z8 Fury still delivered more than enough storage performance for demanding content creation, AI, and professional visualization workflows.

DiskSpeedTest (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Read	8,911.5 MB/s	9,111.4 MB/s
Write	8,166.5 MB/s	9,292.0 MB/s
Secondary Disk Test
Read	4,136.8 MB/s	N/A
Write	5,149.3 MB/s	N/A

3DMark CPU

The 3DMark CPU Profile evaluates processor performance across six threading levels: 1, 2, 4, 8, 16, and max threads. Each test runs the same boid-based simulation workload to assess how well the CPU scales under different thread counts, with minimal GPU involvement. The benchmark helps identify single-threaded efficiency and multithreaded potential for tasks such as gaming, content creation, and rendering. Scores on 8 threads often align with modern DirectX 12 gaming performance, while 1–4-thread results reflect older or esports scenarios.

The 3DMark CPU Profile benchmark showed the HP Z8 Fury G6i delivering solid scaling across thread counts, though it trailed both comparison platforms throughout the test suite. At Max Threads, the HP scored 15,792, finishing about 6.5% behind the Intel Xeon 658x platform (16,890) and 43% behind the AMD-based Dell Precision 7875 (27,670). This trend continued in the lower-thread-count tests, where the HP generally landed within 5–10% of the Xeon system but was further behind the high-core-count Threadripper Pro workstation.

3DMark CPU (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Max Threads	15,792	16,890	27,670
16 Threads	11,241	13,022	15,378
8 Threads	6,635	7,185	8,477
4 Threads	3,594	3,751	4,701
2 Threads	1,816	1,944	2,378
1 Threads	895	990	1,237

3DMark Storage

The 3DMark Storage Benchmark tests your SSD’s gaming performance by measuring tasks like loading games, saving progress, installing game files, and recording gameplay. It evaluates how well your storage performs in real-world gaming and supports the latest storage technologies, providing accurate performance insights.

In the 3DMark Storage Benchmark, the HP Z8 Fury G6i achieved an overall score of 2,944, placing it close to the Dell Precision 7875’s 3,221 result. This left the HP system approximately 8.6% behind the Dell workstation, indicating comparable storage responsiveness for game loading, file transfers, and other storage-intensive workloads. The HP system’s secondary drive also posted a respectable score of 2,067.

3DMark Storage (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Overall Score	2,944	3,221
Overall Score (Secondary Drive)	2,067	N/A

LuxMark

LuxMark is a GPU benchmark that uses LuxRender, an open-source ray-tracing renderer, to evaluate a system’s performance on highly detailed 3D scenes. This benchmark is relevant for assessing the graphical rendering capabilities of servers and workstations, especially for visual effects and architectural visualization applications, where accurate light simulation is crucial.

In LuxMark, the HP Z8 Fury G6i delivered performance very close to that of the Dell Precision 7875. In the Food scene, the HP scored 41,476, trailing Dell’s 41,981 by just 1.2%. The gap widened slightly in the more demanding Hall workload, where the HP reached 95,414 compared to 101,808 from the Precision 7875, a difference of roughly 6.3%. Overall, the results show that the Z8 Fury provides GPU rendering performance nearly equivalent to that of the Dell workstation in ray-traced rendering workloads.

LuxMark (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Food	41,476	41,981
Hall	95,414	101,808

Geekbench 6

Geekbench 6 is a cross-platform benchmark that measures overall system performance.

In Geekbench 6, the HP Z8 Fury G6i delivered performance very similar to the Intel Xeon 658x platform while trailing the AMD-powered Dell Precision 7875. In the CPU tests, the HP scored 2,333 in single-core and 21,110 in multi-core performance, placing it within 2% of the Xeon system while trailing the Precision 7875 by approximately 28% in single-core and 26% in multi-core performance.

On the GPU side, the HP posted 291,727 in OpenCL and 276,201 in Vulkan. Compared to the Dell Precision 7875, which scored 330,765 and 309,146, respectively, the HP trailed by roughly 12% in OpenCL and 11% in Vulkan.

GeekBench (Higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
CPU Single Core	2,333	2,383	3,240
CPU Multi-Core	21,110	21,447	28,618
GPU OpenCL	291,727	N/A	330,765
GPU Vulkan	276,201	N/A	309,146

Y-Cruncher

y-cruncher is a multithreaded and scalable program that can compute Pi and other mathematical constants to trillions of digits. Since its launch in 2009, it has become a popular benchmarking and stress-testing application for overclockers and hardware enthusiasts.

The HP Z8 Fury G6i performed exceptionally well in the Y-Cruncher benchmark, consistently outperforming the Intel Xeon 658x platform and remaining highly competitive with the AMD-based Dell Precision 7875. In the smaller datasets, the HP was the fastest system tested, completing the 250 million-digit run in 1.203 seconds, approximately 30% faster than the Xeon platform and nearly 50% faster than the Precision 7875. This trend continued through the 500-million- and 1-billion-digit tests, with HP maintaining the lead.

As workload sizes increased, the Dell Precision 7875’s larger memory capacity and higher core count began to show their advantage. At 2.5 billion digits, the HP completed the run in 17.0 seconds, trailing the Dell by about 12% while dramatically outperforming the Xeon platform. The gap widened in the larger datasets, with the Precision 7875 leading the 5-, 10-, and 25-billion-digit tests by approximately 21–27%.

Y-Cruncher (lower duration is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
250 Million	1.203 s	1.711 s	2.369 s
500 Million	2.667 s	3.758 s	4.281 s
1 Billion	6.042 s	8.312 s	7.617 s
2.5 Billion	17.047 s	23.065 s	15.188 s
5 Billion	37.890 s	54.207 s	29.795 s
10 Billion	82.983 s	118.898 s	61.572 s
25 Billion	232.832 s	326.454 s	169.289 s
50 Billion	N/A	N/A	371.039 s
100 Billion	N/A	N/A	844.503 s

7-Zip Compression

The 7-Zip Compression Benchmark evaluates CPU performance during compression and decompression, measuring GIPS (Giga Instructions Per Second) and CPU usage. Higher GIPS and efficient CPU usage indicate superior performance.

In the 7-Zip Compression Benchmark, the HP Z8 Fury G6i delivered the strongest overall result among the systems tested. Looking at the Total Rating, the HP achieved 344.964 GIPS, outperforming the Intel Xeon 658x platform’s 250.083 GIPS by approximately 38% and finishing well ahead of the Dell Precision 7875. The HP also led the Resulting Compression Rating, posting 340.875 GIPS compared to 233.557 GIPS from the Xeon platform, a margin of roughly 46%.

The decompression results followed a similar pattern, with the HP reaching a Resulting Decompression Rating of 349.054 GIPS, compared to 266.608 GIPS on the Xeon platform.

7-Zip Compression Benchmark (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Intel Xeon 658x Test Platform (128 GB RAM | NVIDIA RTX 4090)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Compression
Current CPU Usage	5,483%	4,038%	6,445%
Current Rating/Usage	6.247 GIPS	5.786 GIPS	6.949 GIPS
Current Rating	342.522 GIPS	233.641 GIPS	48.392 GIPS
Resulting CPU Usage	5,461%	4,037%	701%
Resulting Rating/Usage	6.242 GIPS	5.705 GIPS	7.010 GIPS
Resulting Rating	340.875 GIPS	233.557 GIPS	49.108 GIPS
Decompression
Current CPU Usage	6,029%	4,688%	728%
Current Rating/Usage	5.839 GIPS	5.705 GIPS	6.801 GIPS
Current Rating	352.023 GIPS	267.475 GIPS	49.526 GIPS
Resulting CPU Usage	5,990%	4,657%	749%
Resulting Rating/Usage	5.827 GIPS	5.725 GIPS	6.832 GIPS
Resulting Rating	349.054 GIPS	266.608 GIPS	51.181 GIPS
Total Rating
Total CPU Usage	5,726%	4,347%	725%
Total Rating/Usage	6.034 GIPS	5.755 GIPS	6.921 GIPS
Total Rating	344.964 GIPS	250.083 GIPS	50.145 GIPS

V-Ray

The V-Ray Benchmark measures rendering performance on CPUs, NVIDIA GPUs, or both, using the advanced V-Ray 6 engines. It uses quick tests and a simple scoring system to help users evaluate and compare their systems’ rendering capabilities. It’s an essential tool for professionals seeking efficient performance insights.

In the V-Ray benchmark, the HP Z8 Fury G6i delivered a score of 28,237, placing it close to the Dell Precision 7875’s 30,356 result. The HP trailed by approximately 7%, indicating that both systems offer similar rendering capabilities for professional visualization and content creation workloads.

V-Ray (higher is better)	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Score	28,237	30,356

SPECworkstation 4.0 Results

The SPECworkstation 4.0 benchmark is a comprehensive tool that evaluates all key aspects of workstation performance. It offers a real-world measure of CPU, graphics, accelerator, and disk performance, ensuring professionals have the data to make informed decisions about their hardware investments. The benchmark includes a dedicated set of tests focused on AI and ML workloads, such as data science tasks and ONNX Runtime-based inference tests, reflecting the growing importance of AI/ML in workstation environments. It encompasses seven industry verticals and four hardware subsystems, providing a detailed and relevant measure of today’s workstations’ performance.

The HP Z8 Fury G6i turned in a strong overall showing in SPECworkstation 4.0, consistently outperforming the Intel Xeon 658x test platform across most workloads while remaining competitive with the AMD-powered Dell Precision 7875. In the industry vertical tests, the HP led the Xeon system in AI & Machine Learning (+11%), Energy (+71%), Financial Services (+90%), Life Sciences (+46%), Media & Entertainment (+9%), and Product Design (+8%), highlighting the benefits of its higher-end workstation configuration and dual professional GPUs.

Within the hardware subsystem scores, the HP maintained advantages over the Xeon platform in CPU performance (+27%), Accelerator performance (+2%), and Graphics performance (+43%), while trailing only in storage performance. Compared to the Dell Precision 7875, the HP generally ranked second, though it remained relatively close in AI & Machine Learning (3.82 vs 4.42), Life Sciences (5.01 vs 5.34), and Accelerator performance (6.27 vs 7.51).

SPECworkstation 4.0	HP Z8 Fury G6i Intel Xeon 696X | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q	Intel Xeon 658X Test Platform 128 GB RAM | NVIDIA RTX 4090	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Industry Vertical Scores
AI & Machine Learning	3.82	3.44	4.42
Energy	6.11	3.57	10.11
Financial Services	5.48	2.89	8.03
Life Sciences	5.01	3.44	5.34
Media & Entertainment	3.61	3.31	4.82
Product Design	2.97	2.75	3.97
Productivity & Development	1.05	1.35	1.72
Hardware Subsystem Scores
CPU	3.21	2.52	4.58
Accelerator	6.27	6.16	7.51
Graphics	11.17	7.82	15.07
Storage	1.03	1.52	1.29

SPECviewperf 15 Results

SPECviewperf 15 is the industry-standard benchmark for evaluating 3D graphics performance across OpenGL, DirectX, and Vulkan APIs. It introduces new workloads, including blender-01 (Blender 3.6), unreal_engine-01 (Unreal Engine 5.4, DirectX 12), and enscape-01 (Enscape 4.0, Vulkan ray tracing), along with updated traces for 3ds Max, CATIA, Creo, Maya, and SolidWorks. With its redesigned GUI, modern application support, and advanced rendering workloads, SPECviewperf 15 provides consistent, real-world insights into professional graphics performance.

In SPECviewperf 15, the HP Z8 Fury G6i delivered strong professional graphics performance and remained competitive with the Dell Precision 7875 despite both systems utilizing dual RTX PRO 6000 GPUs. The HP was particularly strong in Energy and Medical workloads, scoring 116.75 vs. 114.42 and 136.95 vs. 136.06, respectively, giving it a slight advantage in those tests. The two systems were also effectively tied in Blender (90.38 vs. 90.83) and Enscape (52.22 vs. 52.53), with less than a 1% difference between them.

The Dell workstation maintained larger leads in several engineering and CAD-focused workloads, including Creo (+46%), Unreal Engine (+42%), CATIA (+19%), SolidWorks (+16%), and Maya (+17%). However, the HP remained highly competitive across the benchmark suite and demonstrated particularly strong performance in visualization, rendering, and simulation-oriented workloads.

Workload	HP Z8 Fury G6i Intel Xeon 696X | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Composite Scores
3dsmax-08	86.06	92.80
blender-01	90.38	90.83
catia-07	94.77	112.41
creo-04	155.15	227.18
energy-04	116.75	114.42
enscape-01	52.22	52.53
maya-07	137.90	162.04
medical-04	136.95	136.06
snx-05	N/A	93.87
solidworks-08	154.40	178.68
unreal_engine-01	68.35	96.86

Topaz Video AI

Topaz Video AI is a professional application for enhancing and restoring video using advanced AI models. It supports tasks such as upscaling footage to 4K or 8K, sharpening blurry content, reducing noise, improving facial details, colorizing black-and-white footage, and interpolating frames for smoother motion. The suite includes an onboard benchmark that measures system performance across its various video-enhancing algorithms, providing a clear view of how well hardware platforms handle demanding AI video-processing workloads.

In Topaz Video AI, the HP Z8 Fury G6i delivered strong performance across the suite’s video enhancement and upscaling models. However, the Dell Precision 7875 generally maintained the lead in the most demanding AI workloads. In the commonly used 1X enhancement models, the HP reached 37.5 FPS in Artemis, 37.4 FPS in Iris, and 38.5 FPS in Proteus, while the Dell workstation achieved roughly 25–40% higher performance in those same tests. However, the HP did post a notable win in the Gaia model, achieving 16.3 FPS compared to 14.7 FPS on the Dell system.

The trend continued in the heavier 2X and 4X upscaling workloads, where the Dell platform generally delivered higher throughput, reflecting the advantage of its higher-end CPU and larger memory configuration. That said, the HP remained competitive in several motion interpolation tests, outperforming the Dell in 4X Slowmo APFast (52.9 FPS vs. 34.8 FPS) and 4X Slowmo Chronos (37.0 FPS vs. 33.0 FPS).

Overall, the results show the HP Z8 Fury G6i as a capable AI video-processing workstation that performs well across Topaz Video AI’s broad range of enhancement models. At the same time, the Dell Precision 7875 generally leads in the most computationally intensive upscaling and restoration workloads.

Test / Model	HP Z8 Fury G6i (Intel Xeon 696x | 128 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)	Dell Precision 7875 (AMD 9995WX 96C | 512 GB RAM | 2x NVIDIA RTX PRO 6000 Max-Q)
Benchmark Results – 1X
Artemis	37.49 fps	46.87 fps
Iris	37.35 fps	52.63 fps
Proteus	38.49 fps	48.19 fps
Gaia	16.30 fps	14.71 fps
Nyx	15.35 fps	22.90 fps
Nyx Fast	34.83 fps	50.42 fps
Nyx XL	N/A	3.66 fps
Hyperion HDR	21.55 fps	28.57 fps
Benchmark Results – 2X
Artemis	13.45 fps	22.99 fps
Iris	13.24 fps	20.24 fps
Proteus	13.28 fps	23.67 fps
Gaia	11.83 fps	10.36 fps
Nyx	11.89 fps	18.81 fps
Benchmark Results – 4X
Artemis	3.49 fps	6.53 fps
Iris	3.71 fps	6.53 fps
Proteus	3.71 fps	6.40 fps
Gaia	3.61 fps	6.04 fps
Rhea	3.63 fps	5.71 fps
RXL	3.64 fps	5.84 fps
Slow Motion Benchmarks
4X Slowmo – Apollo	30.15 fps	37.51 fps
4X Slowmo – APFast	52.91 fps	34.78 fps
4X Slowmo – Chronos	36.99 fps	33.00 fps
4X Slowmo – CHFast	28.66 fps	38.86 fps
16X Slowmo – Aion	24.69 fps	37.29 fps

HP Z8 Fury G6i vLLM Performance Testing

To evaluate the HP Z8 Fury G6i, we tested configurations using the vLLM Online Serving benchmark, one of the most widely adopted high-throughput inference and serving engines for large language models. The vLLM online serving benchmark simulates real-world production workloads by sending concurrent requests to a running vLLM server and measuring key metrics, including total token throughput (tokens per second), time to first token, and time per output token, under varying load conditions.

Our testing spanned a range of models, from dense architectures to micro-scaling data types. The tests evaluated performance across three workload scenarios: Equal ISL/OSL, Prefill Heavy, and Decode Heavy. These scenarios represent distinct real-world serving patterns, from balanced input and output loads to compute-intensive prompt processing and memory-bandwidth-bound token generation.

To benchmark the HP Z8 Fury G6i, we tested a dual-GPU configuration (2x NVIDIA RTX PRO 6000 Blackwell). Because the system was tested with the same NVIDIA RTX PRO 6000 Blackwell cards used in our Dell Precision 7875 review, the results provide a direct platform-to-platform comparison between the two workstations.

GPT-OSS-120B

Equal ISL/OSL (256/256): The Dell led marginally through batch 32 (4,409 vs 4,341), then the HP edged ahead to 12,604 vs 11,848 tok/s at batch 256 (about 6%).

Prefill Heavy (8k/1k): Nearly identical, HP slightly ahead, climbing to 20,347 vs 18,954 tok/s at batch 256 (about 7%).

Decode Heavy (1k/8k): Virtually tied the entire way, HP peaking at 5,445 vs 5,275 tok/s at batch 256 (about 3%).

GPT-OSS-20B

Equal ISL/OSL (256/256): The Dell led slightly through batch 8 (3,085 vs 3,071), then the HP took over, peaking at 24,131 vs 22,034 tok/s at batch 256 (about 10%).

Prefill Heavy (8k/1k): HP ahead throughout, reaching 35,193 vs 31,982 tok/s at batch 256, the highest absolute throughput in the suite, about 10% over the Dell.

Decode Heavy (1k/8k): Essentially matched through batch 16, HP pulling slightly ahead at scale to 10,461 vs 9,985 tok/s at batch 256 (about 5%).

Qwen3 Coder 30B FP8

Equal ISL/OSL (256/256): The Dell led through the HP’s batch-8 dip (1,783 vs 539), but the HP recovered at batch 16 and pulled steadily away to 18,435 tok/s vs 13,577 at batch 256, a 36% HP advantage.

Prefill Heavy (8k/1k): HP ahead throughout, peaking at 14,780 vs 13,661 tok/s at batch 128 (about 8%); both tapered at batch 256.

Decode Heavy (1k/8k): HP led the full curve, finishing at 3,908 vs 3,464 tok/s at batch 256 (about 13%).

Qwen3 Coder 30B BF16

Equal ISL/OSL (256/256): Even early (the HP took a sharp batch-8 dip to 457), then the HP climbed steeply to 14,282 tok/s vs the Dell’s 10,171 at batch 256, a 40% advantage.

Prefill Heavy (8k/1k): Closely tracked through batch 32, then the HP separated, peaking at 13,485 vs 11,789 tok/s. Both rolled off at batch 256 (HP 12,011 vs 9,381), with HP holding a 28% lead.

Decode Heavy (1k/8k): HP ahead for most of the curve, peaking at 3,409 vs 3,019 tok/s at batch 128 (about 13%).

Mistral Small 24B

Equal ISL/OSL (256/256): Essentially overlapping through batch 64 (4,745 vs 4,730), with the HP edging ahead at the top to 8,766 vs 8,261 tok/s at batch 256 (about 6%).

Prefill Heavy (8k/1k): Tightly matched, both peaking at batch 64 (HP 6,789 vs Dell 6,627) before falling off sharply at higher batches, HP just 2% ahead at peak.

Decode Heavy (1k/8k): Near-identical curves; the Dell even nudged ahead at batch 64, with the HP peaking at 1,894 vs 1,831 tok/s at batch 128 (about 3%). The closest-matched model overall.

Llama 3.1 8B (FP8)

Equal ISL/OSL (256/256): The Dell led early, 427 vs 342 tok/s at batch 1, and held the lead through batch 16 (5,050 vs 2,310), where the HP run dipped at batch 8. The HP recovered hard at batch 32 (8,945 vs 8,341) and pulled away from there, peaking at 23,004 tok/s vs the Dell’s 16,833 at batch 256, a 37% HP advantage.

Prefill Heavy (8k/1k): The two tracked closely, with the Dell marginally ahead at batch 1 (1,803 vs 1,682). The HP took over by batch 4 and stayed ahead, reaching 21,693 tok/s at batch 256 vs 18,822 tok/s, about 15% higher.

Decode Heavy (1k/8k): The HP led throughout by a steady margin, finishing at 6,287 tok/s vs the Dell’s 5,429 at batch 256, a 16% gain.

Llama 3.1 8B BF16

Equal ISL/OSL (256/256): Roughly even through batch 4 (about 1,109 each), with the Dell briefly ahead at batches 8 to 16 during an HP dip. From batch 32 on the HP led, peaking at 17,921 tok/s vs 13,789 at batch 256, 30% higher.

Prefill Heavy (8k/1k): Nearly identical curves, HP slightly ahead the whole way. Both peaked at batch 128 (HP 12,542 vs Dell 11,639), then tapered at batch 256.

Decode Heavy (1k/8k): HP led by a small, consistent margin, peaking at 3,435 vs 3,225 tok/s at batch 128 (about 7%).

Platform Differences: Why the HP Pulls Ahead

Both workstations run identical 2x RTX PRO 6000 Blackwell GPUs, and neither platform supports NVLink, so all inter-GPU communication during tensor-parallel inference travels over PCIe. The Dell Precision 7875 has one PCIe Gen 5 x16 slot with the second GPU running at Gen 4 x16, while the HP Z8 Fury G6i provides full PCIe Gen 5 x16 to all GPU slots.

The host CPU also plays a role since vLLM’s scheduler and token processing run on the CPU rather than the GPU, and the Intel Xeon 6 platform in the HP offers architectural advantages that the Threadripper PRO does not. These platform differences are most pronounced on smaller and quantized models at high concurrency, where the HP leads by 30-37%, and shrink to the low single digits on large MoE models like GPT-OSS-120B, where GPU compute time is the dominant factor.

Conclusion

The HP Z8 Fury G6i is HP’s statement that Intel is back in the high-core workstation conversation. Built around the Granite Rapids-WS Xeon 600 series, our review unit paired the 64-core Xeon 696X with two NVIDIA RTX PRO 6000 Blackwell Max-Q cards, 128 GB of DDR5-6400, and a four-drive NVMe layout, positioning it squarely as an AI workstation rather than a traditional CAD tower. The build reflects HP’s current design language: a uniform matte black chassis with a full-height diamond-mesh grille, paired with a genuinely serviceable interior featuring tool-free NVMe latches, removable dual power supplies, a collapsible rear handle, and a nine-slot, four-PCIe-Gen5-x16 layout that scales to four Max-Q cards.

In our benchmarks, the Z8 Fury staked out clear wins where its platform strengths matter. It dominated the Blackmagic RAW Speed Test at 311 FPS (8K CPU) and 650 FPS (GPU), topped the field in 7-Zip compression at 344.96 GIPS, led the smaller Y-Cruncher datasets, and posted strong SPECworkstation 4.0 results that outpaced the Intel Xeon 658x platform across nearly every vertical. GPU rendering in Blender and LuxMark was effectively a tie with the Dell Precision 7875, as expected given the shared RTX PRO 6000 silicon.

Where it gives ground, the pattern is consistent. The 96-core Threadripper PRO in the Dell still leads in heavily multithreaded and memory-bound workloads, including Blender CPU, the larger Y-Cruncher runs, 3DMark CPU, and several Topaz Video AI models, reflecting its higher core count and 512 GB of memory. The vLLM inference results are where the platform argument gets interesting: running identical dual RTX PRO 6000 cards, the HP pulled ahead by 30 to 40 percent on smaller and quantized models at high concurrency, narrowing to low single digits on large MoE models like GPT-OSS-120B, where GPU compute dominates. With neither platform supporting NVLink, that gap traces back to HP’s full PCIe Gen5 x16 to every GPU slot and the Xeon 6 host handling vLLM’s CPU-side scheduling more effectively than the Threadripper PRO.

At $52,139 as configured, the Z8 Fury G6i is a serious investment, though most buyers will see better volume pricing through corporate channels. What you get is a thoroughly engineered, highly serviceable AI workstation that, depending on workload, trades blows with the best Threadripper PRO towers and pulls clearly ahead in PCIe-bound multi-GPU inference. For organizations standardizing on Intel and prioritizing GPU-accelerated AI work, the HP Z8 Fury G6i makes a strong case for itself.

HP Z8 Fury G6i Product Page

The post HP Z8 Fury G6i Review: One Xeon, up to Four Blackwell GPUs appeared first on StorageReview.com.

As quantum computing continues its transition from research labs to practical computing environments, AMD and IBM are positioning themselves around a common vision: hybrid computing architectures that combine quantum processors, high-performance computing (HPC), and AI into a unified platform. Recent disclosures from both companies outline a growing partnership focused on building the infrastructure required to make quantum computing useful at scale, highlighting a broader industry shift away from standalone quantum systems and toward tightly integrated quantum-classical environments.

The collaboration comes as governments, research institutions, and enterprises increase investment in quantum technologies. The U.S. Department of Commerce announced in May 2026 plans to invest more than $2 billion in quantum computing and quantum manufacturing initiatives, a sign that quantum is increasingly viewed as a strategic capability with implications for scientific discovery, economic competitiveness, and national security. At the same time, industry leaders acknowledge that practical quantum computing will depend as much on advances in classical computing infrastructure as on the quantum processors themselves.

Quantum Computing’s Evolution Toward Hybrid Architectures

A recurring theme throughout AMD’s quantum strategy is that quantum computers are not replacements for traditional systems. Instead, they are specialized accelerators designed to solve classes of problems that remain difficult or impractical for classical computers. As a result, future deployments are expected to rely on hybrid architectures where quantum processors work alongside conventional compute resources.

IBM and AMD describe this model as quantum-centric supercomputing. In such environments, quantum computers are integrated with HPC clusters and AI systems that handle workload orchestration, simulation, optimization, data movement, and preprocessing. Classical computing resources remain responsible for many of the operational functions required to support quantum workloads.

This architectural approach reflects current realities within the quantum ecosystem. While quantum hardware continues to improve, it still requires extensive classical infrastructure for control systems, error mitigation, data processing, and software execution. The result is a computational model in which performance depends on the efficiency of both the quantum and classical layers.

IBM Chairman and CEO Arvind Krishna described quantum computing as a fundamentally new way of representing and processing information, one that enables exploration of computational problems beyond the reach of traditional systems. Through its collaboration with AMD, IBM is exploring how advanced HPC technologies can be combined with quantum systems to create more capable hybrid computing environments.

AMD Chair and CEO Dr. Lisa Su similarly emphasized the importance of HPC as a foundation technology for solving complex scientific and industrial challenges, noting that the convergence of quantum computing and high-performance computing creates opportunities to accelerate research and innovation.

AMD’s Role in the Quantum Ecosystem

Unlike companies developing quantum processors, AMD focuses on providing the classical infrastructure needed to support emerging quantum deployments. The company argues that practical quantum computing will require a broad portfolio of technologies spanning compute, acceleration, networking, and software.

AMD’s quantum-related portfolio includes EPYC processors, Instinct accelerators, Versal adaptive SoCs, Pensando networking technologies, FPGAs, and associated software frameworks. These technologies are intended to provide the computational backbone for quantum environments, supporting everything from system management and simulation to AI-assisted optimization and data processing.

The company has been investing in technologies that support the development of quantum computing for nearly a decade. AMD views its role as enabling the convergence of quantum computing, AI, and HPC rather than competing directly in quantum hardware development.

According to AMD, future quantum deployments will require scalable architectures that integrate multiple computing paradigms into a single environment. The company is working with leaders in scientific research and financial services, including Oak Ridge National Laboratory and JPMorgan Chase, to explore how quantum systems can be integrated into AI and high-performance computing environments.

Open Platforms and Software Integration

Another area of focus for both companies is developing open, scalable software environments that support hybrid quantum workflows.

The partnership aims to leverage IBM’s quantum software expertise, including the open-source Qiskit ecosystem, alongside AMD’s experience in HPC and AI infrastructure. AMD’s ROCm software for HPC is expected to evolve to orchestrate quantum accelerators alongside GPUs, enabling developers and researchers to build applications that span quantum and classical resources without being constrained by proprietary interfaces.

AMD has emphasized its commitment to open software ecosystems and community-developed interfaces as critical elements for long-term quantum adoption. Industry observers generally agree that software interoperability will be a key factor in determining how quickly quantum computing transitions from specialized research environments into broader commercial use.

Industry Implications

The AMD-IBM partnership reflects a growing consensus across the industry that the future of quantum computing will not be defined solely by advances in qubit technology. Instead, value will likely emerge from integrated systems that combine quantum processors with powerful classical infrastructure.

This trend is particularly important for organizations already operating large HPC environments. Rather than replacing existing investments, hybrid quantum architectures are expected to augment traditional systems, enabling enterprises and research institutions to selectively apply quantum resources to specialized workloads while relying on established HPC infrastructure for most processing tasks.

Potential applications include materials science, drug discovery, logistics optimization, financial modeling, and other computationally intensive workloads where quantum algorithms may eventually provide advantages over classical approaches.

Moving Forward

The collaboration between AMD and IBM underscores the industry’s movement toward quantum-enabled supercomputing environments. As quantum technology matures, companies are betting that success will depend on seamless integration between quantum processors, AI accelerators, networking technologies, and traditional HPC infrastructure.

For AMD, the strategy expands its role beyond traditional server and accelerator markets by positioning its compute portfolio as a foundational component of future quantum deployments. For IBM, the partnership provides access to advanced HPC technologies needed to support increasingly sophisticated quantum systems.

While large-scale fault-tolerant quantum computing remains a long-term objective, the work underway today suggests the next phase of the market will be defined by hybrid quantum-classical systems. The AMD-IBM partnership represents one of the more visible examples of how the industry is preparing for that transition, focusing on the infrastructure and software layers required to bring quantum computing into practical production environments.

The quantum computing space is becoming increasingly competitive, with rapid developments such as Microsoft’s work on Majorana-based qubits. If progress continues at this pace, we may see usable quantum computers within the next decade.

The post AMD and IBM Advance Hybrid Quantum-Centric Supercomputing Strategy appeared first on StorageReview.com.

At ISC High Performance 2026 in Hamburg, VDURA is showcasing three major platform updates designed to improve storage operations and data pipeline performance for AI and HPC environments. The company unveiled a next-generation multi-tenant control plane, enhanced S3 performance capabilities, and native S3 object tagging support, all scheduled for general availability in the second half of 2026.

The updates target organizations that manage large-scale AI training, inference, and scientific computing workloads, where operational simplicity and sustained storage performance are critical.

New Control Plane Targets Multi-Tenant AI and HPC Deployments

VDURA’s next-generation control plane introduces a redesigned management experience focused on multi-tenant environments. The platform adds a modernized management interface and a simplified tenant administration model, allowing platform administrators and tenant operators to manage storage resources from a centralized dashboard.

The release also includes a REST API for key platform operations, providing integration points for automation frameworks, orchestration tools, and operational workflows. Together, the interface and API are intended to reduce management complexity while maintaining the level of control required by enterprise, cloud, and research deployments.

S3 Performance Optimized for AI Data Pipelines

VDURA is also delivering a series of S3 performance enhancements aimed at cloud-native AI workflows. The company said the updates are designed to sustain high throughput across data-intensive operations such as model checkpointing, inference serving, and large-scale dataset ingestion.

The improvements focus on reducing latency for S3-native operations while increasing aggregate throughput across concurrent read and write workloads. As AI environments continue to scale in both dataset size and user concurrency, maintaining consistent object storage performance has become increasingly important for minimizing pipeline bottlenecks.

Native S3 Object Tagging Adds Metadata-Based Data Management

Complementing the performance updates, VDURA is introducing native S3 object tagging support. The capability allows organizations to associate metadata with stored objects, creating a framework for policy-driven data management and governance.

With object tagging, administrators can implement lifecycle policies, automate tiering workflows, and apply more granular access controls across datasets, model artifacts, and scientific research repositories. For organizations operating petabyte-scale object storage environments, metadata-driven management can help streamline governance and data lifecycle processes without requiring changes to application workflows.

Customer-Driven Enhancements

“With these new capabilities, we’re focused on making VDURA more powerful and easier to operate at every level of the organization,” said Chris Girard, Vice President of Product Management at VDURA. “The new management interface brings clarity and control to the people running VDURA day-to-day, while our expanded S3 capabilities give data engineers and AI practitioners the tools to build more sophisticated, automated data pipelines. These are the operational and integration capabilities our customers have been asking for.”

Availability

The next-generation control plane, S3 performance enhancements, and native S3 tagging capabilities are expected to become generally available during the second half of 2026 for all V5000-class systems. Existing customers will be able to deploy the new capabilities through an in-place online software update.

The post VDURA Unveils Multi-Tenant Control Plane and S3 Enhancements at ISC 2026 appeared first on StorageReview.com.

Virtuozzo has announced that Virtuozzo Storage has achieved Veeam Ready certification as an object storage backup target and is now listed in the Veeam Ready database. The certification validates interoperability and performance requirements for use with Veeam’s object storage capabilities. It allows customers to use Virtuozzo Storage as a certified backup repository while maintaining existing Veeam backup and recovery workflows.

 

The certification expands backup deployment options for service providers and enterprises already using Veeam. Organizations can integrate Virtuozzo Storage into existing data protection environments without changing operational processes, management tools, or recovery procedures.

For service providers, the certification supports the delivery of Veeam-compatible Backup-as-a-Service offerings on infrastructure they already operate. Because Virtuozzo Infrastructure integrates virtualization, cloud services, AI workloads, and object storage on a single software-defined platform, providers can consolidate backup storage with compute resources rather than maintaining separate storage systems or dedicated backup appliances.

“Veeam has built one of the most trusted ecosystems in data protection, and achieving Veeam Ready status is an important milestone for Virtuozzo,” said Ivan Lukovnikov, Chief Product Officer at Virtuozzo. “Service providers can now offer Veeam-based services on infrastructure they already operate, reducing complexity while creating new revenue opportunities. At the same time, customers gain confidence that their backup data is stored on a certified object storage target supporting data resilience and control as a foundation for Data and AI Trust.”

“Organizations are looking for flexible infrastructure choices that allow them to modernize their environments without changing the data resilience strategies they already rely on,” said Andreas Neufert, Vice President of Product Management, Alliances at Veeam. “Veeam is built for the reality where identity, data, security, and AI are inseparable, and the Veeam Ready program helps customers and partners deploy with confidence. By achieving Veeam Ready status, Virtuozzo has demonstrated interoperability for supported object storage use cases, giving service providers and customers confidence when using Virtuozzo Storage as part of a resilient backup and recovery architecture.”

Virtuozzo Storage was certified as an object storage target for use with the Veeam Data Platform, including:

• Veeam Backup & Replication v13
• Veeam Backup for Microsoft 365 v8
• Veeam Agent for Microsoft Windows v13
• Veeam Agent for Linux v13
• Veeam Agent for Mac v13

Virtuozzo Storage is an S3-compatible object storage platform designed for backup repositories, long-term retention, cloud-native applications, and AI workloads. Integrated into the Virtuozzo Infrastructure System, the platform allows providers to deliver storage and compute services from a common environment, helping reduce infrastructure complexity and operational overhead.

The company positions the platform as an alternative to relying on third-party public cloud storage for backup repositories. By keeping backup data on provider-controlled infrastructure, service providers can maintain greater control over performance, data location, and service delivery. This approach may also support data sovereignty requirements, differentiated disaster recovery services, and recovery service-level objectives by keeping backup repositories closer to production workloads.

The announcement comes as organizations continue to evaluate alternatives to traditional virtualization platforms and seek infrastructure that combines compute, storage, networking, and data protection in a unified architecture. Virtuozzo Infrastructure is designed to address those requirements through a software-defined platform that improves resource utilization, simplifies operations, and reduces overall infrastructure costs.

The post Virtuozzo Storage Achieves Veeam Ready Certification for Object Storage Backups appeared first on StorageReview.com.

CGI and NetApp have expanded their global alliance partnership, with NetApp Keystone set to power CGI’s block storage solutions within its shared services platform. The move deepens the companies’ existing relationship and is designed to help organizations modernize IT infrastructure, improve data management, and advance AI initiatives across private, public, and hybrid cloud environments.

Under the expanded agreement, CGI will integrate NetApp Keystone’s subscription-based storage model into its shared services platform. The offering is designed to provide customers with scalable block storage resources while allowing organizations to consume storage capacity through a flexible, consumption-based model rather than traditional infrastructure procurement cycles.

NetApp Keystone delivers storage services across on-premises and cloud environments, combining performance, data management capabilities, and high-availability features. The platform also incorporates integrated security features focused on threat detection, data protection, and recovery, supporting organizations with increasingly stringent operational and cybersecurity requirements.

The partnership combines NetApp’s storage and data management portfolio with CGI’s consulting, cloud, AI, and managed services expertise. NetApp is one of more than 150 technology firms in CGI’s Global Alliances network, and together the companies are targeting enterprises looking to modernize infrastructure while maintaining operational consistency across hybrid environments.

“The expansion of our partnership with NetApp reflects a strong commitment on both sides to drive meaningful outcomes for our clients,” said Virginia Williams, Senior Vice-President and Business Unit Leader, U.S. Northwest Operations at CGI. “The technology, expertise and innovation offered by this powerful alliance will continue to help clients modernize their IT environments, become more data-driven and prepare for AI at scale.”

This next phase of the alliance will see the two companies working together to design, deliver, and operate solutions to meet clients’ evolving digital needs. CGI will deliver services on behalf of NetApp, while NetApp will partner with CGI to deliver enterprise-grade data and storage services that enable flexible, consumption-based solutions for joint clients across industries.

“By expanding our partnership with CGI, we’re enabling our shared customers to build a resilient, secure solution that delivers consistent performance and intelligent data management for their most critical workloads,” said Alvaro Celis, Chief Partner and Ecosystem Officer at NetApp. “Working side-by-side, CGI and NetApp will continue to empower organizations to achieve better business outcomes through an intelligent data infrastructure that simplifies hybrid cloud adoption and securely unlocks greater value from their data.”

The announcement reflects a broader enterprise trend toward storage-as-a-service consumption models, particularly as organizations seek greater operational flexibility while preparing infrastructure for data-intensive AI workloads and hybrid cloud deployments. By incorporating Keystone into its shared services platform, CGI adds a scalable storage foundation that can be delivered as part of larger managed services and digital transformation engagements.

The post CGI Taps NetApp Keystone to Power Block Storage in Its Shared Services Platform appeared first on StorageReview.com.

IBM has announced the general availability of three new IBM Z software offerings that strengthen security operations, simplify certificate management, and improve database administration through AI-assisted automation. The releases reflect IBM’s continued focus on supporting mission-critical workloads running on IBM Z and LinuxONE environments as enterprises address increasingly complex security, compliance, and operational requirements.

 

IBM positions the new tools alongside its broader security initiatives, including Project Glasswing and Project Lightwell, as organizations face evolving cyber threats and growing regulatory demands. The company noted that enterprises continue to rely on IBM Z for highly sensitive workloads, where platform resiliency and availability remain key requirements.

IBM zSecure Detection

IBM zSecure Detection is designed to improve threat monitoring and incident response on z/OS environments. The software provides continuous visibility into IBM Z activity and is intended to help security teams identify indicators of ransomware, suspicious behavior, and other potential threats across the platform.

Beyond basic monitoring, zSecure Detection combines AI-driven access anomaly detection, network insights, and automated response to flag behaviors such as privilege escalation, unusual dataset access, suspicious command execution, and anomalous cryptographic activity. It also introduces automated network micro-segmentation that builds policies from observed communication patterns to limit lateral movement, and it feeds near-real-time alerts into existing SIEM workflows. IBM said the offering is intended to strengthen the overall security posture as threat actors increasingly target critical enterprise systems.

IBM zSecure Secret Manager

IBM also introduced IBM zSecure Secret Manager, a new solution focused on managing certificates and secrets for IBM Z and LinuxONE deployments.

As certificate lifecycles continue to shorten, organizations face growing operational challenges in tracking, renewing, and managing certificates, a problem set to intensify as the industry moves toward TLS certificate renewal every 47 days. Powered by IBM Vault Self-Managed for Z and LinuxONE as the certificate authority, the solution uses a PKI secrets engine to deliver policy-driven, automated certificate renewal across z/OS environments, helping head off certificate-driven outages caused by expired certificates.

IBM said the software is designed to reduce administrative overhead from fragmented certificate management processes while providing a more centralized approach to secrets and certificate lifecycle management.

IBM Z Database Assistant

The third release, IBM Z Database Assistant, brings agentic AI capabilities to database administration workflows on IBM Z.

Built on a set of DBA-centric AI agents for Db2 and IMS on the mainframe, the software automates routine work such as provisioning, index rebuilds, and log backups, while adding intelligent SQL tuning, deadlock and timeout diagnostics, and continuous object health monitoring. It pairs those agents with context-driven dashboards and natural language interaction. IBM describes the platform as proactive, autonomous, and intelligent, designed to shift database teams from reactive troubleshooting to higher-value work.

The announcement reflects a broader trend of incorporating AI-driven automation into enterprise infrastructure management, particularly in environments where data integrity, availability, and operational consistency are critical.

Supporting Modern IBM Z Operations

The latest additions expand IBM’s software portfolio for IBM Z environments at a time when enterprises are balancing AI adoption, hybrid infrastructure strategies, and increasingly sophisticated cybersecurity threats. With organizations continuing to run core business applications on IBM Z, IBM is focusing on tools that help operations, security, and database teams manage and protect critical workloads while maintaining the platform’s traditionally high availability and resiliency standards.

The post IBM Expands Z Software Portfolio with New Security and Database Management Tools appeared first on StorageReview.com.

AMD announced that its Spartan UltraScale+ SU200P FPGA will enter volume production in July 2026. The SU200P is the largest device in the Spartan UltraScale+ family. It expands AMD’s cost-optimized FPGA portfolio with a combination of high I/O density, low power operation, flexible connectivity, and integrated security features designed for long-lifecycle deployments.

Built on a 16nm FinFET process, the SU200P packs 218,000 system logic cells, 6.8Mb of block RAM, and 18Mb of UltraRAM, up to eight 16.3 Gb/s GTH transceivers, PCIe Gen4 x8 hard IP, and LPDDR4X/LPDDR5/LPDDR5X memory controllers running up to 4,266 Mb/s. AMD targets the device at applications spanning data center infrastructure, networking equipment, industrial automation, and embedded systems, where programmable control, connectivity, and security are increasingly critical.

Scalable FPGA Platform

The SU200P sits at the top of the nine-device Spartan UltraScale+ family and is designed to simplify migration across performance tiers. AMD emphasizes footprint compatibility across the product family, including between the SU65P and SU200P devices, allowing customers to reuse board designs as requirements evolve.

Capability	SU200P FPGA
System logic cells	218,000
1.0V – 3.3V I/O	Up to 572, 3.2 Gb/s MIPI
Block RAM / UltraRAM	6.8 Mb / 18.0 Mb
GTH transceivers	Up to 8 × 16.3 Gb/s
PCIe hard IP	Gen4 x8
Hard memory controllers	LPDDR4X / LPDDR5 / LPDDR5X up to 4,266 Mb/s
Lifecycle	2045

The SBVF900 package also aligns with the Kintex UltraScale+ Gen 2 family, providing a migration path to higher-performance FPGAs without requiring a board redesign. This compatibility can help reduce engineering effort and preserve hardware investments across multiple product generations.

Security Designed for Long-Lifecycle Systems

Security is a major focus of the SU200P, particularly for systems expected to remain deployed for years in environments such as data centers, industrial facilities, and medical infrastructure.

The device incorporates a hardware root of trust and support for post-quantum cryptography aligned with CNSA 2.0 requirements. AMD says the platform is designed to protect device identity, secure firmware authentication, and support trusted updates throughout the operational lifecycle.

Security features include hardware root of trust secure boot, AES-GCM authenticated encryption, dedicated cryptographic engines, a true random number generator compliant with NIST and AIS standards, and a physical unclonable function (PUF) for device-specific identity and key protection. Together, these capabilities establish a chain of trust from manufacturing through field deployment and ongoing firmware maintenance.

Targeting Data Center Infrastructure

AMD positions the SU200P as a companion FPGA for increasingly complex server platforms. Modern servers continue to add sensors, telemetry sources, power domains, and management interfaces, while the growth of AI infrastructure is driving demand for more sophisticated board-level control functions.

The company highlights the FPGA’s potential role as a platform root of trust, enabling device attestation and secure boot capabilities before host processors initialize. The device supports up to 572 I/O pins, PCIe Gen4 connectivity, 3.3V I/O, and programmable control logic that adapts to evolving server architectures and management standards.

AMD also notes support for emerging Open Compute Project technologies such as the Low-speed Transport Protocol Interface (LTPI) and the Management Port Embedded Secure Tunnel Interface (M-PESTI), which are becoming increasingly relevant in next-generation server designs.

Consolidating Functions in Network Switches

For networking applications, the SU200P is designed to integrate control-plane management, interface adaptation, monitoring, and security functions into a single programmable platform.

The FPGA includes PCIe Gen4 connectivity, GTH transceivers, up to two hard memory controllers, and support for the AMD MicroBlaze V soft processor running a real-time operating system. This allows switch vendors to consolidate management and control functions while reducing component count and board complexity.

Security capabilities are integrated into the device’s configuration architecture, enabling secure boot, authenticated firmware updates, hardware-based device authentication, and post-quantum cryptography support throughout the product lifecycle.

Industrial Focus with Extended Availability

Industrial automation, robotics, edge control systems, and sensor aggregation platforms are other key target markets for the SU200P.

AMD has committed to supporting the device through at least 2045, addressing a common requirement in industrial markets where systems often remain deployed for decades. The FPGA integrates up to 18Mb of UltraRAM, GTH transceivers, hard memory controllers, and high I/O density, enabling designers to reduce external components, board area, power consumption, and overall system complexity.

Combined with its integrated security architecture, the device is intended to support long-term deployments that require secure firmware management and a protected device identity over extended operating lifecycles.

Availability

The AMD Spartan UltraScale+ SU200P FPGA enters volume production in July 2026. AMD positions the device as a scalable platform for customers requiring a balance of programmable logic, connectivity, security, and long-term availability across data center, networking, and industrial deployments.

The post AMD Spartan UltraScale+ SU200P FPGA Enters Volume Production appeared first on StorageReview.com.

CloudCasa by Catalogic has introduced CloudCasa Disaster Recovery (CloudCasa DR) for Kubernetes, powered by HPE Alletra Storage MP B10000. The new offering extends CloudCasa’s Kubernetes data protection portfolio with orchestrated disaster recovery capabilities designed for Kubernetes applications and KubeVirt virtual machines deployed across HPE and hybrid cloud environments.

The announcement builds on CloudCasa’s existing support for backup and recovery of HPE Kubernetes Service (HKS) clusters. With the addition of CloudCasa DR, organizations can leverage storage-based replication and automated recovery workflows to meet more aggressive recovery objectives than traditional backup-based approaches.

Disaster Recovery Driven by Native Replication

Rather than restoring workloads from backups after an outage, CloudCasa DR uses native replication built into the HPE Alletra Storage MP B10000 platform. The approach is intended to deliver near-zero data loss and faster recovery times for business-critical Kubernetes environments.

CloudCasa combines Kubernetes-aware orchestration with array-native replication to automate failover and failback operations between clusters and sites. This enables organizations to recover workloads with minimal manual intervention while maintaining application consistency during disaster recovery events.

Support for Kubernetes Applications and Virtual Machines

The platform protects both containerized applications and KubeVirt-based virtual machines, including deployments running on Red Hat OpenShift Virtualization and SUSE Virtualization environments.

Additional capabilities include bidirectional disaster recovery, namespace-level recovery, multi-tenant isolation, role-based access controls, and enterprise security features. Supported Kubernetes platforms include HPE Kubernetes Service, Red Hat OpenShift, SUSE Rancher, and other Kubernetes distributions.

Extending CloudCasa and HPE Collaboration

CloudCasa positions its offering to help organizations meet enterprise RPO and RTO targets without the delays of full backup restores. “Organizations running Kubernetes on HPE infrastructure need disaster recovery that meets enterprise RPO and RTO targets without the delay of full backup restores,” said Ryan Kaw, vice president of global sales and alliances at CloudCasa. “By combining Kubernetes-aware orchestration with HPE Alletra Storage MP B10000 native replication, CloudCasa helps customers recover mission-critical applications and virtual machines in minutes across hybrid and multi-cluster environments.”

The launch further expands the collaboration between CloudCasa and HPE, providing customers with a broader data protection strategy that includes backup, recovery, and disaster recovery for Kubernetes workloads. Organizations can also use HPE StoreOnce and HPE X10000 Object Storage as backup targets for long-term retention and recovery operations.

With CloudCasa DR, the companies are targeting enterprises seeking automated disaster recovery capabilities for Kubernetes infrastructure while maintaining support for hybrid cloud deployments and modern virtualization workloads.

The post CloudCasa Adds Kubernetes Disaster Recovery for HPE Alletra Storage MP B10000 appeared first on StorageReview.com.

The HighPoint Rocket 1604L is a $399 PCIe Gen5 x16 add-in card that carries four M.2 NVMe SSDs, each on a dedicated Gen5 x4 connection. In our testing with four Samsung 9100 PRO 4TB drives installed, the card sustained 55.6GB/s of 128K sequential read bandwidth and 10.1 million 4K random write IOPS, numbers that are within a few percent of what the four drives are rated to deliver on native motherboard slots. That is the entire pitch of this card: it adds drive bays without subtracting performance.

The 1604L takes a different architectural path than most quad-M.2 cards we have looked at. It is not a passive bifurcation riser, nor is it a PCIe switch card. Instead, it is built around an Astera Labs PT5161LRS retimer, which sits in the data path at the physical layer, re-clocking and regenerating the Gen5 signal between the host slot and each M.2 connector. At Gen4 speeds, passive cards that simply route traces from the slot to the connectors are usually fine. At Gen5’s 32GT/s signaling rate, trace length and connector transitions start eating into the signal budget, and marginal links train down to Gen4 or throw correctable errors under load. The retimer approach addresses that without the cost, power, and latency of a full PCIe switch. The trade-off is that the host platform must support x4/x4/x4/x4 bifurcation on the slot, since the retimer does not perform any lane virtualization of its own.

This card joins a HighPoint Gen5 family we have covered previously, which includes the switch-based Rocket 1604A, which works in any x16 slot regardless of bifurcation support, and the Rocket 7604A, which adds bootable RAID on top. The 1604L is the leanest of the three. There is no RAID stack and no driver; the operating system simply enumerates four native NVMe devices, and anything beyond that (mdadm, Storage Spaces, ZFS) is up to the user. HighPoint positions the card heavily toward servers hosting M.2 accelerator modules like the Hailo-8 series, but for our purposes, the storage use case is the more universal one. The card is a full-height, half-length design that HighPoint claims is roughly 40% shorter than typical four-bay M.2 cards, with a full-length anodized aluminum heatsink, thermal padding for the drives, an integrated low-decibel fan, and a ventilated bracket. Firmware-level monitoring exposes per-port lane allocation, power draw, and board health, with present and activity LEDs for each SSD.

The bifurcation requirement is the caveat to settle before buying. Most mainstream consumer boards either cannot split a x16 slot four ways or steal those lanes from the primary GPU slot. Where the 1604L makes immediate sense is on platforms with PCIe lanes to spare: Threadripper TRX50 and WRX90, Xeon W, and EPYC or Xeon server boards, where x4/x4/x4/x4 is a BIOS toggle and a spare x16 slot is not a sacrifice. That describes our test rig, so the fit was natural.

HighPoint Rocket 1604L Specifications

Specification	Rocket 1604L (R1604L)
Bus Interface	PCIe 5.0 x16
Chipset	Astera Labs PT5161LRS retimer
Working Mode	4 x 4-lane (host bifurcation x4/x4/x4/x4 required)
Ports	4x M.2 NVMe (dedicated PCIe 5.0 x4 per port)
Device Support	M.2 NVMe SSDs or M.2 PCIe accelerator modules
SSD Form Factors	M.2 2242, 2260, 2280
Data Transfer Rate	Up to 64GB/s
RAID Support	None (OS-level software RAID optional)
Form Factor	Full-height, half-length
Cooling	Full-length aluminum heatsink, integrated fan, thermal pads, ventilated bracket
Monitoring	Per-port lane allocation, power, and health via smart firmware; present and activity LEDs
OS Support	Native NVMe support in mainstream operating systems, x86 Intel/AMD and ARM
Price	$399 (HighPoint eStore)

Build and Design

The 1604L’s compact footprint is the visible difference from the sprawling four-bay cards of the Gen4 era. Drive installation is conventional: heatsink off, drives into the four sockets, thermal pads aligned, heatsink back on. The single fan exhausts through the ventilated bracket, which matters in workstation towers where slot airflow is unpredictable. We did not observe thermal throttling from any of the four drives during sustained 60-second test runs.

Testing Setup

We tested the Rocket 1604L in our consumer Threadripper platform, the same water-cooled rig that has handled our recent high-end GPU and HEDT CPU reviews. The card was installed in a Gen5 x16 slot configured for x4/x4/x4/x4 bifurcation.

StorageReview Threadripper Test Platform

• CPU: AMD Ryzen Threadripper 7980X (64C/128T)
• Motherboard: ASUS Pro WS TRX50-SAGE WIFI
• RAM: 128GB DDR5-6400
• Storage: 1TB Gen4 Boot SSD, 4x Samsung 9100 PRO 4TB (FW 0B2QNXH7) on the Rocket 1604L
• OS: Ubuntu Server 24.04

The four Samsung 9100 PRO drives are each rated at 14,800MB/s sequential read, 13,400MB/s sequential write, 2,200K random read IOPS, and 2,600K random write IOPS, which puts the theoretical aggregate at 59.2GB/s read and 8.8 million random read IOPS. Since a Gen5 x16 slot tops out at roughly 63GB/s of usable bandwidth, the drives, not the slot, are the ceiling in this configuration. That is the right way around; a card like this should never be the bottleneck.

All workloads were run with FIO 3.36 using the io_uring engine against the raw block devices, with a 5% LBA span per drive, 60-second runtimes with a 5-second ramp, and one job per drive at QD64 for sequential transfers or 16 jobs per drive at QD32 (64 total) for 4K random. These are burst-oriented consumer test parameters rather than enterprise steady-state methodology, consistent with how we evaluate client platform accessories.

HighPoint Rocket 1604L Performance

Sequential Bandwidth

Workload (4 drives aggregate)	IOPS	Bandwidth	Avg Latency	99th % Latency
128K Sequential Read, QD64	424K	55.6GB/s	604µs	906µs
128K Sequential Write, QD64	279K	36.5GB/s	918µs	1,303µs
64K Sequential Read, QD64	668K	43.8GB/s	383µs	570µs
64K Sequential Write, QD64	462K	30.3GB/s	553µs	914µs

The headline number is the 128K sequential read result of 55.6GB/s, which works out to 13.9GB/s per drive, or about 94% of Samsung’s 14,800MB/s rating for the 9100 PRO. Getting four Gen5 drives to within striking distance of their individual spec sheets, simultaneously, through a single add-in card is the result that validates the retimer architecture. Average latency held at 604µs with the 99th percentile at 906µs, and per-drive utilization stayed pinned above 99% for the duration of the run. The 64K read result of 43.8GB/s trails the 128K figure as expected, since larger transfers amortize protocol overhead more efficiently.

Sequential writes landed at 36.5GB/s at 128K and 30.3GB/s at 64K. That is below the four drives’ combined 53.6GB/s write rating, which is a drive behavior rather than a card limitation: vendor write specs reflect short bursts into pSLC cache, while our 60-second sustained runs push past that window. The write latency profile stayed orderly, with the 128K test averaging 918µs and holding 1,303µs at the 99th percentile.

4K Random Performance

Workload (4 drives aggregate)	IOPS	Bandwidth	Avg Latency	99th % Latency
4K Random Read, QD32 x 64 jobs	8.83M	36.2GB/s	231µs	553µs
4K Random Write, QD32 x 64 jobs	10.1M	41.5GB/s	202µs	461µs

The random results are the cleanest evidence that the 1604L’s data path is transparent. Samsung rates the 9100 PRO 4TB at 2,200K random read IOPS, and four of them behind the 1604L produced 8.83 million, which is the rated aggregate almost to the decimal. Random write reached 10.1 million IOPS against a theoretical ceiling of 10.4 million, about 97% of spec. Writes-outrunning-reads looks odd at first glance but matches the drives’ own ratings, helped along by the 5% working set, which keeps the controllers operating in their happiest caching range.

Latency under these loads stayed tight, averaging 231µs for reads and 202µs for writes, with 99th percentile figures of 553µs and 461µs, respectively. The other observation worth passing along is host cost: driving nearly 10 million IOPS through 64 FIO jobs consumed roughly 60% of the system CPU time over the run. The card will hand a workstation more storage performance than most applications can absorb, and feeding it is a workload in its own right.

Conclusion

The Rocket 1604L does one job, and our test data shows it doing that job with effectively no overhead. Four Samsung 9100 PRO 4TB drives delivered 55.6GB/s of sequential read bandwidth, 8.83 million random read IOPS, and 10.1 million random write IOPS through the card, figures that sit at 94 to 100% of the drives’ combined ratings. For a device whose value proposition is invisibility, that is a clean sweep.

The buyer’s question is whether the $399 ask is justified, given that passive bifurcation cards sell for a fraction of that price. At Gen4 and below, it often is not. At Gen5, the signal integrity margin is thin enough that the retimer earns its keep, particularly for users planning to load the card with drives that each move 14GB/s. Worked out per bay, $100 per Gen5 M.2 slot with cooling and monitoring included is reasonable against the alternative of unstable link training on a passive card, and it undercuts switch-based options while preserving the full bandwidth of every port.

Who should buy it: TRX50, WRX90, Xeon W, and server platform owners who want 16TB or more of Gen5 flash in a single slot for media work, AI dataset staging, or scratch space, and who are comfortable with OS-level RAID or none at all. Who should not: anyone on a platform without x4/x4/x4/x4 bifurcation support, who should look at the switch-based Rocket 1604A instead, and anyone needing bootable hardware RAID, which is the Rocket 7604A’s territory. Buyers running Gen4 drives can also save money with simpler cards, since the retimer’s advantages are largely wasted below 32GT/s.

HighPoint Rocket 1604L Product Page

The post HighPoint Rocket 1604L Review: Four Gen5 M.2 SSDs, One Slot, 55.6GB/s appeared first on StorageReview.com.

CTERA has announced a new integration with n8n, the agentic workflow automation platform, to connect enterprise file data with AI services, applications, and business processes. The integration extends the CTERA Intelligent Data Platform into the rapidly growing ecosystem of agentic AI and workflow automation tools while maintaining enterprise governance and security controls.

The integration introduces native CTERA community nodes for n8n, enabling organizations to build workflows to securely search, access, and manage data stored within a CTERA enterprise data fabric. The platform spans edge locations, corporate sites, and cloud environments, providing a centralized framework for accessing distributed file data.

Bringing Enterprise File Data into AI Workflows

A key objective of the integration is to make unstructured enterprise content more accessible to AI-powered automation. Organizations can securely connect file data managed by CTERA to hundreds of applications, AI services, databases, and business systems within the n8n ecosystem.

CTERA’s platform classifies file data and adds contextual understanding, turning traditional file repositories into data sources that AI agents and automated workflows can act on. Rather than relying solely on file events or basic triggers, workflows can draw on content’s meaning, classification, metadata, and compliance status.

Content-Aware Automation

The integration enables n8n workflows to incorporate CTERA Search, CTERA Classify, and CTERA Experts capabilities. This allows workflows to make decisions based on file content, metadata, compliance classifications, and business context.

By incorporating content-aware decision-making into automation pipelines, organizations can create workflows that go beyond basic task orchestration. Automated processes can evaluate compliance requirements, understand document context, and act on enterprise content with greater accuracy.

Simplifying Workflow Development

CTERA also highlights the ability to accelerate automation projects using n8n’s visual workflow designer. Organizations can create integrations and business processes without extensive custom development, reducing implementation complexity while leveraging governed enterprise data sources.

An early adopter, Bezeq Group, one of Israel’s largest telecommunications providers, is exploring the integration across its hundreds of terabytes of CTERA-managed file data. “Bezeq differentiates itself with customer service, and we believe that AI and automation will play a key role in supporting our customers and helping improve and streamline internal workflows across all departments,” said Igal Muginstein, Storage Team Manager in Bezeq’s Engineering & Network Division, citing the company’s interest in pairing n8n with CTERA’s data classification.

The combined platform is intended to support a range of enterprise use cases, including compliance-driven document routing, AI-assisted knowledge management, file lifecycle automation, secure collaboration workflows, and integration with broader business processes.

Focus on Production AI Deployments

CTERA CEO Oded Nagel said organizations are increasingly moving from AI experimentation to production deployments, creating a greater need for trusted enterprise data sources. He noted that the n8n integration allows organizations to connect automation and AI initiatives directly to governed enterprise content while maintaining security and operational controls.

According to CTERA, the integration is designed to help organizations transform enterprise storage from a passive repository into a platform that actively supports automation, AI-driven workflows, and business decision-making.

The post CTERA Integrates with n8n to Extend Enterprise File Data to AI Workflows appeared first on StorageReview.com.

Sandisk has expanded its Optimus gaming SSD lineup with the new SANDISK Optimus GX PRO 850P NVMe SSD for PS5 consoles, alongside the SANDISK Optimus GX 7100X NVMe SSD for ROG Xbox Ally X and PC. The company also announced availability for several other Optimus drives.

Sandisk Optimus GX PRO 850P NVMe SSD

The Sandisk Optimus GX PRO 850P NVMe SSD is officially licensed for PlayStation 5 and PlayStation 5 Pro, with testing and certification for Sony’s console platform. It also features an exclusive heatsink design with PlayStation branding, which is built for the PS5 M.2 slot, so buyers do not need to add a separate heatsink.

With game install sizes, updates, and DLC continuing to eat into console storage, the Optimus GX PRO 850P is designed for users who want to keep more titles installed and ready to play. Sandisk lists capacities up to 8TB, giving PS5 owners enough room for much larger game libraries while reducing the need to delete older titles to make space for new releases. The drive supports playing games directly from the SSD once installed, which makes it a direct expansion option rather than just a place to store inactive games.

Performance is based on PCIe Gen 4.0 NVMe technology, with Sandisk quoting sequential speeds of up to 7,300MB/s read and 6,600MB/s write, depending on capacity. For the 1TB model, Sandisk lists up to 7,300MB/s sequential read and 6,300 MB/s sequential write, along with 800K random read IOPS and 1.1M random write IOPS. The 1TB model is also rated for 600TBW endurance and carries a five-year limited warranty.

Although the 850P is mainly marketed as a PS5 and PS5 Pro upgrade, Sandisk also lists compatibility with computers that have an M.2 M-key slot and support the M.2 2280 form factor, along with Windows 10 and newer.

Sandisk Optimus GX 7100X NVMe SSD

The Sandisk Optimus GX 7100X NVMe SSD gives ROG Xbox Ally, ROG Xbox Ally X, and PC users an officially licensed storage upgrade built for portable gaming and larger game libraries. It supports capacities up to 4TB, giving players more room for Xbox titles, updates, DLC, and Game Pass downloads without constantly managing installs. Sandisk also includes a one-month Xbox Game Pass Ultimate trial in the box.

Performance comes from a PCIe 4.0 NVMe interface, with Sandisk listing sequential read and write speeds of up to 7,250 MB/s and 6,900 MB/s on the 2TB model. The 2TB model is also rated for up to 1M random read IOPS and 1.4M random write IOPS, uses the M.2 2280 form factor, measures 3.15 x 0.87 x 0.09 inches, and features a five-year limited warranty.

The Optimus GX 7100X is also power-efficient, which is especially relevant for handheld gaming PCs, where power draw can affect battery life. The drive is also built with SanDisk’s 8th-generation BiCS TLC 3D CBA NAND and is tested for ROG Xbox Ally, ROG Xbox Ally X, and PC use.

Metric/Field	SANDISK Optimus GX PRO 850P NVMe SSD	SANDISK Optimus GX 7100X NVMe SSD
Overview
Product Name	SANDISK Optimus GX PRO 850P NVMe SSD for PS5 consoles	SANDISK Optimus GX 7100X NVMe SSD for ROG XBOX Ally X and PC
Positioning	Officially licensed for PlayStation 5 and PlayStation 5 Pro consoles	Officially licensed storage for ROG XBOX Ally, ROG XBOX Ally X, and PC
Maximum Capacity	Up to 8TB	Up to 4TB
Form Factor	M.2 2280	M.2 2280
Performance
Interface	PCIe 4.0 NVMe	PCIe 4.0 NVMe
Sequential Read Performance	7,300MB/s	7,250MB/s
Sequential Write Performance	6,300MB/s	6,900MB/s
Maximum Sequential Read/Write Speeds	Up to 7,300/6,600 MB/s	Up to 7,250/6,900 MB/s
Random Read	800K IOPS	1M IOPS
Random Write	1.1M IOPS	1.4M IOPS
Hardware and Design
Heatsink	Integrated heatsink Exclusive heatsink design featuring the PlayStation logo Optimized for the PlayStation 5 and PlayStation 5 Pro consoles’ M.2 slot	Not specified
NAND	Not specified	Sandisk’s 8th generation BiCS TLC 3D CBA NAND
Power Efficiency	Not specified	Designed for power efficiency for low-power consumption for laptops and ROG XBOX Ally X
Physical Specifications
Dimensions (L x W x H)	3.15″ x 0.96″ x 0.39″	3.15″ x 0.87″ x 0.09″
Weight	30.4gms	Not specified
Reliability
Warranty	5-Year Limited Warranty	5-Year Limited Warranty
Endurance (TBW)	600	1,200
Operating Temperature	0°C to 85°C	N/A
Non-Operating Temperature	-40°C to 85°C	N/A
Compatibility
Primary Compatibility	PlayStation 5 and PlayStation 5 Pro	ROG XBOX Ally, ROG XBOX Ally X, and PC
PC Compatibility	Computers with M.2 (M-key) port (Capable of taking M.2 2280 form factor) Windows 10+	PC laptops ROG XBOX Ally ROG XBOX Ally X
Product Features
Features	Experience high-speed gaming SSD with PCIe 4.0 technology. New SSD heatsink design specifically built for PS5 and PS5 Pro consoles. Download and play games directly off the drive.	Equipped with PCIe 4.0 interface provides the speed and power for on-the-go XBOX gaming. Designed for power efficiency for low-power consumption for laptops and ROG XBOX Ally X. Endurance of up to 2,400 TBW.
Included Offer	N/A	1-month trial of XBOX Game Pass Ultimate inside the box
Model and Availability
Model Number	SDSG81100TAH-000E0	SDSG71200TAN-000G0
Starting Price	$474.99	$799.99
Availability	Sandisk store and select retailers	Sandisk store and select retailers

Availability and Pricing

The Sandisk Optimus GX PRO 850P NVMe SSD for PS5 consoles is available now through the Sandisk store and select retailers, with pricing starting at $474.99.

Pricing for the Sandisk Optimus GX 7100X NVMe SSD starts at $799.99, with availability now through the Sandisk store and select retailers.

Other Releases

Alongside these launches, the broader Sandisk Optimus lineup is now available through the Sandisk store and select retailers, including:

• SANDISK Optimus GX PRO 8100 NVMe SSD: Designed for professionals, gamers, and creators, the GX PRO 8100 is positioned as a high-performance PCIe 5.0 drive for demanding AI workflows, intensive gaming, and creative workloads. Pricing starts at $524.99.
• SANDISK Optimus GX PRO 850X NVMe SSD: Built for users who need high-capacity storage for gaming and creative applications, the GX PRO 850X offers capacities up to 8TB for larger game libraries, applications, and project files. Pricing starts at $488.99.
• SANDISK Optimus GX 7100 NVMe SSD: The GX 7100 is designed for laptops and handheld gaming consoles, using a power-efficient architecture for gaming sessions and creative workflows on the move. Pricing starts at $207.99.
• SANDISK Optimus GX 7100M NVMe SSD: Built for portable systems, the GX 7100M supports upgrades for compatible Steam Deck, MSI Claw, Microsoft Surface, and Dell laptop systems, with capacities up to 2TB for modern AAA games. Pricing starts at $387.99.
• SANDISK Optimus 5110 NVMe SSD: Planned for release later this year, the Optimus 5110 targets creators seeking faster application launches, greater capacity, and more room for high-resolution video and image files.

The post Sandisk Expands Optimus SSD Lineup with New PS5 and ROG Xbox Ally Storage Options appeared first on StorageReview.com.

AMD and Rackspace Technology have signed a definitive agreement to deploy an initial 30MW footprint of AMD-based compute capacity across Rackspace’s global data center portfolio. The phased rollout is scheduled to begin in late 2026 and continue through 2028, advancing the memorandum of understanding the companies announced in May.

Under the agreement, AMD becomes a strategic silicon partner within Rackspace’s enterprise AI infrastructure strategy. The deployment will combine AMD Instinct accelerators, including the MI355X, MI350P, and future generations, with AMD EPYC processors to support AI training, inference, and enterprise workloads in regulated industries.

At full deployment, the 30MW environment is expected to provide substantial AI compute capacity for enterprise customers, including organizations in healthcare and other regulated sectors. Rackspace said the infrastructure is designed to support large-scale clinical AI initiatives, inference services, and other workloads that require governance, accountability, and operational oversight.

The companies plan to integrate the AMD hardware stack into Rackspace’s Enterprise AI Cloud architecture. The platform is intended to match workloads with the appropriate compute resources while providing centralized management and operational accountability across the infrastructure stack.

Rackspace CEO Gajen Kandiah said regulated industries require AI infrastructure that is governed end-to-end rather than assembled from multiple independent providers. He positioned the collaboration as an effort to combine compute infrastructure and operating services into a single managed framework with accountability extending from the hardware layer through business outcomes.

AMD Senior Vice President and General Manager of Compute and Enterprise AI Dan McNamara said enterprise AI deployments increasingly require a mix of accelerated and general-purpose computing resources optimized for different workload requirements. He noted that the combination of AMD’s AI compute portfolio and Rackspace’s managed cloud operating model is intended to provide enterprises with scalable and accountable infrastructure for production AI environments.

The agreement also includes joint go-to-market activities. Both companies will dedicate sales and marketing resources and commit personnel to jointly develop and pursue customer opportunities across regulated industries built on AMD-powered infrastructure.

The deployment is expected to accelerate the delivery of the four services outlined in the earlier memorandum of understanding:

• Enterprise AI Cloud
• Enterprise Inference Engine
• Inference as a Service
• Bare Metal AMD Instinct

Together, these offerings are designed to provide a managed AI infrastructure stack spanning bare-metal compute through fully operated inference services. The two companies position the initiative as an alternative to traditional self-managed bare-metal AI deployments, targeting enterprises moving beyond pilot projects and into production AI and agentic workflow implementations within core business systems.

The post AMD and Rackspace Formalize 30MW AI Infrastructure Deployment Across Global Data Centers appeared first on StorageReview.com.

The Kioxia CD9P-R is the read-intensive arm of the company’s new data center NVMe SSD generation, and the first CD-series drive built on BiCS FLASH generation 8 TLC. The series pairs Kioxia’s own controller and firmware with PCIe 5.0 and NVMe 2.0, with rated performance reaching 14,800 MB/s sequential read and 2.6 million random read IOPS depending on capacity. E3.S models run from 1.92TB to 30.72TB, while the 2.5-inch variant extends the stack to 61.44TB. Our review unit is the 7.68TB E3.S model in SED trim (KCD9DPJE7T68).

The capacity stack has a wrinkle worth understanding before the charts. Kioxia’s published specs peak in the middle of the family, not at the top: the 7.68TB and 15.36TB models carry the line’s best sequential read rating at 14,800 MB/s and its best random write rating at 450K IOPS, while the 30.72TB flagship steps down to 13,500 MB/s and 270K IOPS. The two smallest capacities also ship on the prior BiCS generation 5 NAND rather than generation 8. In other words, the 7.68TB drive on our bench is the configuration where this platform shows its full hand, and buyers chasing maximum density at 30.72TB give some of that back, which is not uncommon.

The generational step from the CD8P is where Kioxia is making its case, and the published spec tables back it up. At 7.68TB, the outgoing CD8P-R E3.S rated at 200K random write IOPS; the CD9P-R lifts that to 450K, a 2.25x improvement. Sequential read climbs 23% from 12,000 MB/s to 14,800 MB/s, and random read moves 30% from 2 million to 2.6 million IOPS. The active power rating rises slightly, from 21W typical to 23W at this capacity, putting the CD9P-R in line with the rest of the Gen5 read-intensive class. The efficiency argument here concerns what the drive delivers within that envelope.

The fine print that defines where this drive belongs: the CD9P-R features a single-port, 1 DWPD design. There is no dual-port path for traditional enterprise storage arrays, and write-heavy workloads belong on the CD9P-V mixed-use sibling. This is a drive built for hyperscale and cloud server fleets, OLTP read tiers, content delivery, and virtualized environments where the access pattern is read-dominated, and the power budget is fixed. It checks the expected platform boxes along the way, with OCP Datacenter NVMe SSD v2.5 support (not all requirements), power loss protection, end-to-end data protection, SIE and SED security options, a 2.5 million-hour MTTF at 50°C, and a five-year warranty.

KIOXIA CD9P-R Specifications

The table below outlines the KIOXIA CD9P-R Series in the E3.S form factor across its capacity points, highlighting performance metrics, endurance ratings, power, and reliability specifications.

KIOXIA CD9P-R Series Specifications (E3.S)
	30.72TB	15.36TB	7.68TB	3.84TB	1.92TB
Model Numbers
SIE Model Number	KCD9XPJE30T7	KCD9XPJE15T3	KCD9XPJE7T68	KCD9XPJE3T84	KCD9XPJE1T92
SED Model Number	KCD9DPJE30T7	KCD9DPJE15T3	KCD9DPJE7T68	KCD9DPJE3T84	KCD9DPJE1T92
Basic Specifications
Use Case	Read Intensive (1 Drive Write Per Day)
Form Factor	E3.S, 7.5mm thickness
Interface / Protocol	PCIe 5.0 x4, NVMe 2.0
Maximum Interface Speed	128 GT/s (PCIe Gen5 x4)
NAND	KIOXIA BiCS FLASH 3D TLC (Gen 8 for 7.68TB-30.72TB; Gen 5 for 1.92TB-3.84TB)
OCP Compliance	OCP Datacenter NVMe SSD Specification v2.5 (partial)
Security	SIE (Sanitize Instant Erase), SED (TCG Opal & Ruby SSC)
Performance (Up To)
Sequential Read (128KiB, MB/s)	13,500	14,800	14,800	14,500	14,500
Sequential Write (128KiB, MB/s)	7,000	7,000	7,000	7,000	3,600
Random Read (4KiB, K IOPS)	2,600	2,600	2,600	2,600	2,000
Random Write (4KiB, K IOPS)	270	450	450	320	160
Power Requirements
Supply Voltage	12V ±10%, 3.3V ±15%
Power (Active)	23W typ.
Power (Ready/Idle)	5W typ.
Reliability
MTTF	2,500,000 hours @ 0–50°C | 2,000,000 hours @ 0–55°C
UBER	< 1 sector per 1017 bits read
DWPD	1
Warranty	5 Years
Data Protection	Power Loss Protection (PLP), End-to-End Data Protection
Dimensions
Thickness	7.5mm +0.2 / -0.5mm
Width	76mm ±0.25mm
Length	112.75mm ±0.4mm
Weight	110g max
Environmental
Temperature (Operating)	0°C to 75°C
Temperature (Non-operating)	-40°C to 85°C
Humidity (Operating)	5% to 95% RH
Vibration (Operating)	21.27 m/s² { 2.17 Grms } (5–800 Hz)
Shock (Operating)	9.8 km/s² { 1,000 G } (0.5 ms)

KIOXIA CD9P-R Performance

Drive Testing Platform

We use a Dell PowerEdge R760 running Ubuntu 22.04.2 LTS as our test platform for all workloads in this review. Equipped with a Serial Cables Gen5 JBOF, it offers wide compatibility with U.2, E1.S, E3.S, and M.2 SSDs. Our system configuration is outlined below:

• 2 x Intel Xeon Gold 6430 (32-Core, 2.1GHz)
• 16 x 64GB DDR5-4400
• 480GB Dell BOSS SSD
• Serial Cables Gen5 JBOF
• NVIDIA L4

Drives Compared

• Pascari X200P 7.68TB
• SanDisk SN861 7.68TB
• Solidigm PS1010 7.68TB
• Kingston DC3000ME 7.68TB
• Micron 7600 Max 6.4TB
• Micron 9550 MAX 12.8TB
• Micron 9550 Pro 7.68TB

DLIO Checkpointing Benchmark

To evaluate SSD real-world performance in AI training environments, we utilized the Data and Learning Input/Output (DLIO) benchmark tool. Developed by Argonne National Laboratory, DLIO is specifically designed to test I/O patterns in deep learning workloads. It provides insights into how storage systems handle challenges such as checkpointing, data ingestion, and model training.

The chart below illustrates how the drives handle the process across 18 checkpoints. When training machine learning models, checkpoints are essential for periodically saving the model’s state, preventing loss of progress during interruptions or power failures. This storage demand requires robust performance, especially under sustained or intensive workloads. We used DLIO benchmark version 2.0 from the August 13, 2024, release.

To ensure our benchmarking reflected real-world scenarios, we based our testing on the LLAMA 3.1 405B model architecture. We implemented checkpointing using torch.save() to capture model parameters, optimizer states, and layer states. Our setup simulated an eight-GPU system, implementing a hybrid parallelism strategy with 4-way tensor parallelism and 2-way pipeline parallel processing distributed across the eight GPUs. This configuration yielded a checkpoint size of 1,636GB, reflecting the requirements of training modern large language models.

Looking at the pass averages, the KIOXIA CD9P-R started at 464.7 seconds in Pass 1 before increasing to 575.6 seconds in Pass 2 and settling at 572.2 seconds in Pass 3. This behavior closely mirrored the majority of the comparison group, which clustered between roughly 553 and 590 seconds by the final pass. The standout outlier was the Pascari X200P, which finished substantially higher at 674.5 seconds. Overall, the CD9P-R demonstrated predictable scaling across repeated checkpoint operations and remained competitive with the mainstream enterprise Gen5 SSDs in the test.

For the DLIO Checkpoint Benchmark through checkpoint 12, the KIOXIA CD9P-R 7.68TB remained one of the more consistent drives in the comparison group. After starting at 471.4 seconds at the first checkpoint, it settled into a relatively narrow operating range of roughly 560 to 580 seconds for the remainder of the test, finishing checkpoint 12 at 569.7 seconds. The KIOXIA drive closely tracked the Solidigm PS1010, Micron 7600 MAX, and Kingston DC3000ME throughout most of the workload.

The Pascari X200P was the clear outlier, jumping sharply after checkpoint 4 and remaining well above the field, reaching nearly 690 seconds by checkpoint 12. The Micron 9550 MAX showed the lowest sustained checkpoint times during the latter half of the run, dipping as low as 531.3 seconds before ending at 569.1 seconds. While the CD9P-R was not the fastest drive at any individual checkpoint, it avoided the large swings seen from several competitors and delivered stable checkpoint performance across the full test window.

FIO Performance Benchmark

To measure the storage performance of each SSD across common industry metrics, we leverage FIO. Each drive undergoes the same testing process, which includes a preconditioning step of two full drive fills with a sequential write workload, followed by steady-state performance measurement. As each workload type being measured changes, we run another preconditioning fill of that new transfer size.

In this section, we focus on the following FIO benchmarks:

• 128K Sequential
• 64K Random
• 16K Random
• 4K Random

128K Sequential Write (IODepth 16 / NumJobs 1)

Moving to the steady-state 128K Sequential Write test at a lower IODepth of 16, the overall group ranking remained largely unchanged compared to preconditioning. The Micron 9550 Max (12.8TB) continued to lead at 10,957.9 MB/s, with the Micron 9550 Pro (7.68TB) close behind at 10,354.6 MB/s. The Kingston DC3000ME (7.68TB) held third at 8,477.4 MB/s, and the Pascari X200P (7.68TB) was right behind at 8,369.7 MB/s.

The KIOXIA CD9P-R (7.68TB) delivered 6,912.4 MB/s, landing at the back of the field. The Solidigm PS1010 (7.68TB) at 7,126.5 MB/s and the SanDisk DC SN861 (7.68TB) at 7,116.5 MB/s both trailed the mid-pack drives, but still edged out the CD9P-R and the Micron 7600 Max (6.4TB) at 6,960.6 MB/s. The KIOXIA result is consistent and predictable for a read-optimized NVMe drive.

128K Sequential Write Latency (IODepth 16 / NumJobs 1)

At an IODepth of 16 for the steady-state write test, latency dropped substantially across all drives compared to preconditioning conditions. The Micron 9550 Max (12.8TB) again led with the lowest mean latency at 182.2 µs, comfortably ahead of the Micron 9550 Pro (7.68TB) at 192.9 µs, with both drives benefiting from their higher write throughput to service IOs more efficiently.

The KIOXIA CD9P-R (7.68TB) posted 289.0 µs, the highest write latency in the group at this queue depth. The Solidigm PS1010 (7.68TB) at 280.3 µs and the SanDisk DC SN861 (7.68TB) at 280.7 µs were just below the CD9P-R, while the Micron 7600 Max (6.4TB) came in at 287.1 µs. The Kingston DC3000ME (7.68TB) and Pascari X200P (7.68TB) occupied the middle tier at 235.6 µs and 238.6 µs, respectively.

128K Sequential Read (IODepth 64 / NumJobs 1)

The 128K Sequential Read test produced a complete reversal of the write rankings, and the KIOXIA CD9P-R (7.68TB) came out as one of the top performers in the group. The CD9P-R delivered 14,235.9 MB/s, effectively tying with the Pascari X200P (7.68TB) at 14,242.1 MB/s at the top of the chart. The Solidigm PS1010 (7.68TB) at 14,163.3 MB/s, the Micron 9550 Pro (7.68TB) at 14,050.1 MB/s, and the Micron 9550 Max (12.8TB) at 14,047.5 MB/s all clustered tightly in a 200 MB/s band at the top.

The Kingston DC3000ME (7.68TB) trailed the leaders at 13,513.8 MB/s, and the SanDisk DC SN861 (7.68TB) came in at 12,631.2 MB/s. The Micron 7600 Max (6.4TB) at 11,240.5 MB/s was the only drive to fall below the 12 GB/s threshold.

128K Sequential Read latency (IODepth 64 / NumJobs 1)

The 128K Sequential Read latency results closely mirror the bandwidth outcome. The Pascari X200P (7.68TB) led with 561.4 µs, with the KIOXIA CD9P-R (7.68TB) essentially matched at 561.7 µs. The Solidigm PS1010 (7.68TB) at 564.5 µs, Micron 9550 Pro (7.68TB) at 569.0 µs, and Micron 9550 Max (12.8TB) at 569.1 µs all fell within an 8 µs window of the leader, confirming that this tier of drives is constrained by Gen5 interface bandwidth rather than internal latency.

The Kingston DC3000ME (7.68TB) followed at 591.6 µs and the SanDisk DC SN861 (7.68TB) at 633.0 µs, while the Micron 7600 Max (6.4TB) at 711.4 µs posted latency that was 26% higher than the top performers, consistent with its lower sequential read throughput.

 

64K Random Write

Across the full 64K Random Write sweep, the KIOXIA CD9P-R (7.68TB) delivered a consistently respectable bandwidth profile, averaging in the 3-6 GB/s range and reaching a peak of 6,906 MB/s at the highest queue depths tested (IODepth 32 / NumJobs 8). This positioned the CD9P-R in the middle of the field for 64K write throughput, clearly behind the Micron 9550 Max (12.8TB), which scaled to 10+ GB/s peaks, but ahead of the Solidigm PS1010 (7.68TB) and SanDisk DC SN861 (7.68TB), which lagged in the lower half of the chart. The Micron 7600 Max (6.4TB) tracked closely, reaching a similar ceiling and ending just above the CD9P-R.

64K Random Write Latency

The 64K Random Write latency sweep for the KIOXIA CD9P-R (7.68TB) showed us a fairly balanced drive. At low queue depths, latency was well-controlled, starting in the sub-100 µs range at IODepth 1 / NumJobs 1. As concurrency increased, latency rose gradually across the 300-700 µs range for most of the mid-depth sweep, then climbed further at peak queue depths, reaching into the low 2,000 µs range. This placed the CD9P-R in the middle of the group for most of the sweep, performing more predictably than the Solidigm PS1010 (7.68TB) and Pascari X200P (7.68TB), which had sharper spikes in the 4,000-6,000 µs range at high concurrency.

The Micron 9550 Max (12.8TB) maintained the most consistent latency across the sweep, rarely exceeding 1,700 µs even at peak depths, while the Micron 7600 Max (6.4TB) and Micron 9550 Pro (7.68TB) tracked nearby.

64K Random Read

The 64K Random Read sweep revealed one of the more distinctive benefits of the KIOXIA CD9P-R (7.68TB): exceptional performance at low queue depths. At the low end of the sweep, IODepth 1 / NumJobs 1, the CD9P-R started at approximately 1,334 MB/s, leading by a wide margin thanks to its extremely low per-IO read latency. This advantage persisted across the lower queue-depth range, as it consistently ran at or near the top.

As queue depth increased into the 8/4-32/4 range, other drives caught up and surpassed the CD9P-R. At the highest concurrency levels, the CD9P-R stabilized around 11.0-12.0 GB/s, placing it behind the Pascari X200P (7.68TB), Micron 9550 Pro (7.68TB), and Micron 9550 Max (12.8TB), which reached 13.5-14.2 GB/s.

64K Random Read Latency

The 64K Random Read latency sweep confirmed the KIOXIA CD9P-R’s (7.68TB) read latency advantage at low queue depths. From IODepth 1 / NumJobs 1 through the low-concurrency range, the CD9P-R consistently posted some of the tightest latency in the group, tracking well below most competitors and within a narrow band through mid-queue depths.

As queue depth increased beyond the 32/1-to-16/4 range, the group converged, and by the end of the sweep, all drives had climbed into the 600 µs to 1,400 µs range.

16K Random Write

Across the 16K Random Write IOPS sweep, the KIOXIA CD9P-R (7.68TB) delivered a consistent performance profile through the mid-range of the sweep. The Micron 9550 Max (12.8TB) again dominated, sustaining a highly elevated IOPS trajectory well above the field, often approaching 600-690K IOPS at higher queue depths, while the Micron 7600 Max (6.4TB) maintained strong throughput in the 400-450K range.

The CD9P-R tracked alongside the Kingston DC3000ME (7.68TB), Micron 9550 Pro (7.68TB), and SanDisk DC SN861 (7.68TB) through much of the sweep, generally ranging between 200-250K IOPS at moderate depths and scaling toward 365-440K IOPS at the highest tested concurrency (IODepth 32 / NumJobs 16). The Solidigm PS1010 (7.68TB) was the weakest performer in 16K random write, frequently trailing all other drives.

16K Random Write Latency

In the 16K Random Write latency sweep, the KIOXIA CD9P-R (7.68TB) consistently tracked in the lower-to-middle tier across most of the queue depth range. At low queue depths, the drive began under 50 µs and maintained stable, well-controlled latency through the moderate portion of the sweep. As queue depth increased into the 8/8-32/8 range, latency climbed more steeply across all drives, and the CD9P-R moved through the 500-750 µs range before peaking at approximately 1,165 µs.

The Micron 9550 Max (12.8TB) posted the most stable latency across the sweep, holding below the field at most data points. The Solidigm PS1010 (7.68TB) and Pascari X200P (7.68TB) showed the most pronounced latency spikes at high queue depths, peaking at 3,300 µs and 2,050 µs, respectively. In contrast, the CD9P-R, Micron 7600 Max (6.4TB), and Kingston DC3000ME (7.68TB) tracked more predictably through the upper concurrency range. The CD9P-R’s 16K write latency is one of the most consistent in the group under heavy mixed-parallelism loads.

 

 

16K Random Read

In the 16K Random Read IOPS sweep, the Pascari X200P (7.68TB) and Micron 9550 Max (12.8TB) posted the highest sustained read IOPS, approaching 900K at saturation. The Micron 9550 Pro (7.68TB) followed closely with similar scaling, while the Solidigm PS1010 (7.68TB) rounded out the top tier.

The KIOXIA CD9P-R (7.68TB) delivered 734.3K IOPS at its peak measured point (IODepth 32 / NumJobs 8), placing it in the upper-middle tier, ahead of the Micron 7600 Max (6.4TB) at 719.3K, Kingston DC3000ME (7.68TB) at 665.6K, and SanDisk DC SN861 (7.68TB) at 661.1K IOPS.

16K Random Read Latency

In the 16K Random Read latency sweep, the KIOXIA CD9P-R (7.68TB) showed excellent read latency across the majority of the tested queue depth range. From the low end of the sweep, the CD9P-R began at approximately 33 µs at IODepth 1 / NumJobs 1. It remained tightly grouped with the top performers through the moderate-concurrency portion of the sweep, generally staying below 100 µs.

As queue depth increased into the highest concurrency ranges, all drives saw latency climb sharply. The CD9P-R scaled to approximately 713 µs at the peak, landing behind the SanDisk DC SN861 (7.68TB) and Kingston DC3000ME (7.68TB), which exceeded 820-845 µs, while the Micron 9550 Max (12.8TB) and Micron 9550 Pro (7.68TB) maintained a lower peak latency.

4K Random Write

In the 4K Random Write IOPS sweep, the Micron 7600 Max (6.4TB) and Micron 9550 Max (12.8TB) held the top positions through most of the sweep, sustaining 1,500-1,780K IOPS at peak concurrency. The Micron 9550 Pro (7.68TB) and Pascari X200P (7.68TB) followed in the upper tier, while the Solidigm PS1010 (7.68TB) and SanDisk DC SN861 (7.68TB) traded positions in the mid-upper range.

The KIOXIA CD9P-R (7.68TB) reached 1,263.1K IOPS at IODepth 32 / NumJobs 16, placing it in the lower-middle tier for 4K random write performance.

4K Random Write Latency

The 4K Random Write latency sweep produced one of the more variable pictures in the group, and the KIOXIA CD9P-R (7.68TB) again occupied the middle ground. At low queue depths, the CD9P-R’s write latency started in the 9-13 µs range at IODepth 1, competitive but slightly above the best performers at those settings. Through the moderate portion of the sweep, the drive held in the 20-120 µs range alongside most of the group.

At the highest queue depths, the CD9P-R climbed to 200-410 µs, staying well below the Solidigm PS1010 (7.68TB), which peaked near 740 µs and showed significant volatility, and the Pascari X200P (7.68TB) at 544 µs. The Micron 9550 Max (12.8TB), Micron 7600 Max (6.4TB), and Micron 9550 Pro (7.68TB) posted the most controlled latency profiles across the full sweep.

4K Random Read

The 4K Random Read sweep delivered one of the more interesting results in our FIO tests for the KIOXIA CD9P-R (7.68TB). At the lowest queue depths, IODepth 1 / NumJobs 1 and IODepth 2 / NumJobs 1, the CD9P-R separated itself clearly from the field, producing approximately 32.3K IOPS (126 MB/s). As concurrency increased, the rest of the group caught up, and by the peak of the sweep, the rankings shifted. The SanDisk DC SN861 (7.68TB) led in concurrency with 2,555.6K IOPS. At the same time, the KIOXIA CD9P-R reached 2,165.0K IOPS at IODepth 32 / NumJobs 16, effectively tying with the Micron 9550 Max (12.8TB) at 2,164.8K IOPS for second place in the group. The Solidigm PS1010 (7.68TB) and Pascari X200P (7.68TB) followed, with the Micron 7600 Max (6.4TB) and Kingston DC3000ME (7.68TB) at the lower end.

4K Random Read Latency

The 4K Random Read latency sweep confirmed what the IOPS chart suggested: the KIOXIA CD9P-R (7.68TB) has the lowest measured read latency at low queue depths among the drives we tested in the comparison group. Starting at approximately 30 µs with IODepth 1 / NumJobs 1, the CD9P-R had nearly half the latency of most competing drives, clustered in the 60-90 µs range at the same data point. Through the moderate portion of the sweep, the CD9P-R maintained its latency lead, tracking below the group through IODepth 8 and 16 before converging with the pack as concurrency climbed. At the highest tested queue depths, all drives moved into the 120-275 µs range, with the SanDisk DC SN861 (7.68TB) posting the highest latency at roughly 200 µs. The KIOXIA CD9P-R ended the sweep at approximately 236 µs, putting it midway in the group.

GPU Direct Storage

One of the tests we conducted on this testbench was the Magnum IO GPU Direct Storage (GDS) test. GDS is a feature developed by NVIDIA that allows GPUs to bypass the CPU when accessing data stored on NVMe drives or other high-speed storage devices. Instead of routing data through the CPU and system memory, GDS enables direct communication between the GPU and the storage device, significantly reducing latency and improving data throughput.

How GPU Direct Storage Works

Traditionally, when a GPU processes data stored on an NVMe drive, the data must first travel through the CPU and system memory before reaching the GPU. This process introduces bottlenecks, as the CPU acts as an intermediary, adding latency and consuming valuable system resources. GPU Direct Storage eliminates this inefficiency by enabling the GPU to access data directly from the storage device via the PCIe bus. This direct path reduces data-movement overhead, enabling faster, more efficient data transfers.

AI workloads, especially those involving deep learning, are highly data-intensive. Training large neural networks requires processing terabytes of data, and any delay in data transfer can lead to underutilized GPUs and longer training times. GPU Direct Storage addresses this challenge by ensuring that data is delivered to the GPU as quickly as possible, minimizing idle time and maximizing computational efficiency.

In addition, GDS is particularly beneficial for workloads that involve streaming large datasets, such as video processing, natural language processing, or real-time inference. By reducing the reliance on the CPU, GDS accelerates data movement and frees up CPU resources for other tasks, further enhancing overall system performance.

GDSIO Sequential Read Throughput

The GDSIO Sequential Read Throughput sweep highlighted the strong read performance of the KIOXIA CD9P-R (7.68TB). Across all three block size segments, the CD9P-R emerged as the highest-throughput drive at most thread counts.

In the 16K block-size segment, the CD9P-R opened at approximately 0.3 GiB/s on 16K/1 thread, starting behind most of the field. Drives like the Pascari X200P (7.68TB) and Micron 7600 Max (6.4TB) entered the segment at roughly 0.55-0.6 GiB/s on a single thread. As the thread count increased, the CD9P-R scaled aggressively, pulling ahead of the group through the mid-range. By 16K/16 it had overtaken most competitors, and by 16K/64 it reached approximately 2.0 GiB/s, the highest throughput in the group at that point. At 16K/128, the CD9P-R held at roughly 1.9 GiB/s and remained among the top performers as several other drives plateaued or declined. The CD9P-R’s 16K read throughput advantage is a product of how efficiently it scales with thread count, not single-stream dominance.

Moving into the 128K block-size segment, the CD9P-R opened at approximately 1.6 GiB/s at 128K/1, near the top of the group at that thread count, alongside the SanDisk DC SN861 (7.68TB). It scaled cleanly through the moderate thread range, reaching approximately 3.9 GiB/s at 128K/8, the highest throughput in the group at that point, where most competitors were still in the 1.7 to 2.0 GiB/s range. Through 128K/16 and 128K/32, the CD9P-R continued leading the field as other drives scaled up. By 128K/64 and 128K/128, the Micron 9550 Pro (7.68TB) and Micron 9550 Max (12.8TB) caught and passed the CD9P-R, reaching approximately 5.2 to 5.3 GiB/s, while the CD9P-R peaked at approximately 5.1 GiB/s and then pulled back to roughly 4.8 GiB/s at 128K/128.

In the 1M block-size segment, the CD9P-R opened at approximately 3.9 GiB/s at 1M/1 and scaled steadily with each subsequent thread count. At 1M/128, it reached approximately 6.2 GiB/s, the highest peak read throughput in the comparison group. The Pascari X200P (7.68TB) and Micron 9550 Max (12.8TB) were close behind at approximately 6.1 GiB/s each, followed by the Solidigm PS1010 (7.68TB) and Micron 9550 Pro (7.68TB) at roughly 6.0 GiB/s. The Kingston DC3000ME (7.68TB) reached approximately 5.9 GiB/s, while the Micron 7600 Max (6.4TB) was the clear laggard at approximately 5.6 GiB/s.

GDSIO Sequential Read Latency

The GDSIO Sequential Read Latency chart showed that all eight drives followed nearly identical trajectories throughout the full sweep. Latency values are dominated by block size and thread count rather than drive-specific characteristics, making this chart less differentiating than the throughput and IOPS results.

At the single-thread baseline (16K block, 1 thread), the CD9P-R posted approximately 44 µs, placing it in the middle of the group. Faster drives at this configuration included the Pascari X200P (7.68TB) at approximately 26 µs, the SanDisk DC SN861 (7.68TB) at approximately 26 µs, and the Micron 7600 Max (6.4TB) at approximately 27 µs. At the other end, the Kingston DC3000ME (7.68TB) posted approximately 83 µs, and the Solidigm PS1010 (7.68TB) approximately 71 µs, by far the widest single-thread read latency outliers in the group.

Through the rest of the sweep, the lines converged. All drives reached the 20,000-22,000 µs range by 1M/128, with the CD9P-R at approximately 20.3 ms landing in the lower portion of the group, consistent with its higher throughput at that configuration.

GDSIO Sequential Write Throughput

The GDSIO Sequential Write Throughput sweep revealed notable differences across the group.

The Solidigm PS1010 (7.68TB) exhibited a dramatic collapse in write throughput at high thread counts in the 128K block-size segment. After delivering a competitive 3.9 GiB/s at 128K/32, the drive fell sharply to approximately 2.5 GiB/s at 128K/64 and then to just 1.6 GiB/s at 128K/128, less than half the throughput of most competing drives at that configuration. The degradation continued into the 1M segment, where the drive’s peak reached only 4.2 GiB/s with moderate thread counts, then declined again at the high end. This sustained collapse under heavy multi-thread write load stands out as the most significant anomaly in the dataset. The Micron 9550 Max (12.8TB) also showed a sharp throughput drop in the 1M segment, falling from approximately 4.7 GiB/s at 1M/32 to roughly 2.2 GiB/s at 1M/64 before a partial recovery at 1M/128.

In the 16K block-size segment, all drives tracked closely within the 0.6 to 1.5 GiB/s range, with differences of only a few tenths of GiB/s across most thread counts. The CD9P-R opened at approximately 0.7 GiB/s at 16K/1, scaled to approximately 1.5 GiB/s at 16K/32, and then leveled off. The Micron 9550 Max (12.8TB) was the outlier, dropping abruptly to approximately 0.7 GiB/s at 16K/128 while the rest of the group held in the 1.2 to 1.4 GiB/s range.

In the 128K segment, the CD9P-R was among the strongest performers at low thread counts. At 128K/1, it entered at approximately 2.7 GiB/s near the top of the group, and scaled to approximately 4.7 GiB/s at 128K/8, the highest in the group at that point. From 128K/16 onward, the Micron 9550 Pro (7.68TB) and Micron 9550 Max (12.8TB) scaled past the CD9P-R, reaching approximately 5.1 to 5.3 GiB/s at their respective peaks. The CD9P-R held at roughly 4.1-4.5 GiB/s for the remainder of the 128K segment.

In the 1M-block-size segment, the CD9P-R entered the segment at approximately 4.9 GiB/s at 1M/1, the highest throughput in the group at that single-thread configuration. Still, it remained essentially flat at approximately 4.3-4.4 GiB/s across all subsequent thread counts, rather than continuing to scale. Most other drives used additional threads to climb further: the Micron 9550 Pro (7.68TB), Micron 7600 Max (6.4TB), and Pascari X200P (7.68TB) all scaled past the CD9P-R at 1M/4 and higher, reaching peak throughput of approximately 5.2 to 5.7 GiB/s. The CD9P-R’s peak write throughput of 4.9 GiB/s ranked fifth among the eight drives, behind the Micron 9550 Max at 5.69 GiB/s, Micron 9550 Pro at 5.54 GiB/s, Micron 7600 Max at 5.44 GiB/s, and Pascari X200P at 5.41 GiB/s.

GDSIO Sequential Write Latency

In the 16K segment, all drives tracked closely at low thread counts. At 16K/1, the CD9P-R posted approximately 21 µs, the lowest write latency in the group at that configuration, just below the Solidigm PS1010 (7.68TB) at approximately 22 µs. The Pascari X200P (7.68TB) followed at approximately 25 µs, while the Micron 9550 Max (12.8TB) and Micron 7600 Max (6.4TB) both came in around 30 µs. At 16K/128, the Micron 9550 Max spiked to approximately 2,500 µs, the highest write latency in the group at that configuration, while the CD9P-R held at approximately 1,500 µs and remained among the lower values in the group.

In the 128K segment, the Solidigm PS1010’s throughput collapse appeared in the latency chart as a sharp spike to approximately 9,700 µs at 128K/64, the most visible anomaly in that portion of the sweep. The CD9P-R tracked with the main group throughout the 128K segment, remaining in the 1,900 to 3,800 µs range across 128K/64 and 128K/128.

In the 1M segment, two drives showed elevated latency spikes at 1M/64 consistent with their throughput degradation at that configuration. The Micron 9550 Max (12.8TB) and Micron 9550 Pro (7.68TB) each reached approximately 27,000-28,000 µs at 1M/64, then climbed further at 1M/128. The CD9P-R reached approximately 14,500 µs at 1M/64, reflecting its more stable write throughput in that configuration, and ended the sweep at approximately 29,100 µs at 1M/128. The Pascari X200P (7.68TB) delivered the lowest write latency at 1M/128 at approximately 25,000 µs. The highest values at 1M/128 were observed for the Micron 9550 Pro at approximately 44,900 µs and the Solidigm PS1010 at approximately 40,700 µs, with the Micron 7600 Max following at approximately 36,000 µs.

Conclusion

The KIOXIA CD9P-R E3.S 7.68TB does exactly what a read-intensive drive should, and the test data tracks that brief from start to finish. Sequential reads landed near the top of the group at 14,235.9 MB/s in our 128K test, effectively tying the Pascari X200P, and the drive’s real signature showed up at low queue depths: roughly 30 µs of 4K random read latency at QD1, close to half the 60 to 90 µs posted by most of the field. That low-latency read behavior carried over to GPU Direct Storage, where the CD9P-R scaled cleanly with thread count and achieved the group’s highest 1M read throughput of approximately 6.2 GiB/s.

The trade-offs are just as clear and align with the drive’s 1 DWPD design rather than working against it. Write performance sat mid-pack, with 128K sequential writes of 6,912.4 MB/s, landing at the back of the group, and GDSIO sequential write throughput peaking at 4.9 GiB/s, placing fifth among the eight drives. The Micron 9550 family led those write workloads in both throughput and latency. None of that is a knock on the CD9P-R; it is a read-intensive SSD, and buyers who need sustained write performance should be looking at the mixed-use CD9P-V instead.

Consistency under load is the other part of the story. In DLIO checkpointing against the LLAMA 3.1 405B profile, the CD9P-R settled around 572 seconds per pass and stayed inside the 553 to 590 second band that defined the mainstream Gen5 field, avoiding the swings that pushed the Pascari X200P out to 674.5 seconds. The platform also scales well beyond our sample: the CD9P-R family runs to 30.72TB in E3.S and 61.44TB in the 2.5-inch form factor on the same architecture, which lets a single qualification cover everything from performance compute nodes to high-density read tiers.

For cloud fleets, AI data pipelines, content delivery, and virtualized read tiers with read-dominated access patterns, the CD9P-R is an easy drive to recommend. It pairs top-tier Gen5 read throughput with the lowest low-QD read latency in our comparison group and a deep capacity stack, and it holds steady under sustained load.

Product Page – KIOXIA CD9P-R 7.68TB

The post Kioxia CD9P-R Review: Read-Intensive Gen5 Up to 61.44TB appeared first on StorageReview.com.

At HPE Discover 2026, HPE announced a broad set of GreenLake enhancements to help enterprises modernize hybrid infrastructure, manage AI operations, and reduce virtualization complexity. The updates span agentic AIOps, private cloud platforms, virtualization alternatives, and software for AI infrastructure management.

The announcements reflect HPE’s continued push to position GreenLake as a unified operating model for hybrid cloud and AI environments, providing centralized operations, governance, and automation across infrastructure, applications, and AI workloads.

“As enterprises scale AI, they need a simpler way to govern AI infrastructure and modernize operations across hybrid environments without fragmentation or unpredictable costs,” said Fidelma Russo, executive vice president, Hybrid Cloud and CTO at HPE. “The latest advancements in GreenLake give enterprises a proven, unified path for agentic hybrid operations today and foundation for future autonomous operations.”

GreenLake Intelligence Adds Agentic AI Operations

At the center of the updates is GreenLake Intelligence, HPE’s agentic AI framework for hybrid cloud and AI operations.

The platform introduces centralized agent management through an agent registry, orchestration capabilities, and governance controls designed to coordinate AI agents across infrastructure, applications, and operational workflows.

HPE is also expanding HPE OpsRamp with a new Operations Copilot that provides visibility into AI agents and large language models. The platform enables organizations to monitor AI utilization, track token consumption, and understand operational costs across AI factories and hybrid infrastructure environments.

Using telemetry correlation and AI-driven root cause analysis, OpsRamp Operations Copilot can proactively identify operational issues and accelerate troubleshooting workflows.

HPE also announced a partnership with ServiceNow to integrate GreenLake Intelligence and OpsRamp observability capabilities into ServiceNow’s AI-driven service management platform. The integration is intended to create a common operational framework spanning infrastructure monitoring and autonomous service delivery.

Morpheus Gains AI-Driven Automation and Centralized Management

HPE continues to position Morpheus as a virtualization and private cloud platform for organizations seeking alternatives to traditional virtualization environments.

The latest release adds HPE Morpheus Orchestration Copilot, a GreenLake Intelligence capability that automates infrastructure and workload provisioning through AI-assisted workflows. The platform supports a bring-your-own-model approach while applying governance and security controls to orchestration processes.

HPE also introduced HPE Morpheus Central, providing centralized governance and management across multiple Morpheus deployments through a single interface.

Several previously announced capabilities are now generally available. Software-defined networking support brings multitenancy, zero-trust security controls, policy enforcement, and VXLAN overlay networking into Morpheus environments while reducing provisioning time by up to 60%.

The platform also now supports intent-based network automation through integration with HPE Juniper Apstra. The capability continuously validates network configurations, detects drift, and automates policy enforcement.

In addition, stretched cluster functionality is now generally available, enabling active-active deployments across two sites with synchronous replication and automated failover for higher availability.

HPE Zerto integration further supports virtualization modernization efforts by enabling live workload migration from VMware environments to HPE virtual machines while maintaining continuous data protection.

New Programs Target Virtualization Migrations

Following announcements at the HPE Partner Growth Summit 2026, HPE is introducing additional programs to accelerate virtualization migrations.

A new platform migration program lets new HPE Morpheus VM Essentials customers receive up to one free year of VM Essentials licenses and a year of HPE Zerto for $1 to support non-disruptive migration to HPE virtual machines, with 0% interest on software financing through HPE Financial Services. The program is intended to reduce migration costs and help organizations avoid double-paying for overlapping virtualization licenses during transitions.

For service providers, HPE introduced HPE CloudOps Software, a platform designed to support the delivery of private cloud services. The software includes multitenancy, self-service provisioning, software-defined networking, policy-based governance, and cost management capabilities.

The offering is paired with HPE’s Cloud Commit model, which provides pricing and service benefits tied to committed spending levels.

Private Cloud Portfolio Gains Air-Gapped Enhancements

HPE also announced updates across its Private Cloud portfolio focused on operational consistency from edge environments through core data centers.

HPE Private Cloud PC3000 now supports standardized air-gapped deployments for disconnected and regulated environments. The platform also adds validation for VMware vSphere 9, enabling customers to stay current with VMware infrastructure while using Morpheus to manage virtual machines and containers through a common control plane.

HPE Private Cloud PC7000 receives similar VMware vSphere 9 validation while incorporating the latest Morpheus capabilities, including Terraform support, infrastructure-as-code workflows, and automated private cloud operations.

For government and highly regulated deployments, the air-gapped version of PC7000 now supports Department of Defense Impact Level 4 (IL4) certification requirements. HPE said the enhancements address secure design, configuration hardening, vulnerability management, and compliance objectives commonly required in sovereign and regulated environments.

GreenLake Flex Expands Hybrid Infrastructure Management

HPE is also updating GreenLake Flex Solutions with additional operational and procurement capabilities.

A new integrated management interface combines infrastructure observability, sustainability metrics, and consumption analytics into a single operational view. The goal is to simplify hybrid infrastructure management while providing better visibility into resource utilization and costs.

HPE also announced that customers can now purchase selected third-party software offerings directly through the GreenLake Marketplace, extending the platform’s ecosystem capabilities.

Combined with GreenLake’s consumption-based pricing model, the updates are intended to give organizations greater flexibility in managing infrastructure investments while supporting modernization initiatives across hybrid cloud and AI environments.

Availability

HPE OpsRamp Operations Copilot within GreenLake Intelligence is available today, as are HPE CloudOps Software for cloud service providers and the GreenLake Marketplace, which supports direct customer-to-ISV transactions.

The latest HPE Morpheus Software updates are rolling out across the second and third quarters of 2026. Air-gapped deployments of HPE Private Cloud PC3000 and PC7000, along with additional Private Cloud capabilities, are expected in the third quarter of 2026.

GreenLake Intelligence and ServiceNow integrations will roll out across 2026 and 2027.

The post HPE Expands GreenLake for Agentic Operations, Private Cloud, and Virtualization Modernization appeared first on StorageReview.com.

Everpure has announced the availability of Everpure Data Stream, a new platform component based on the NVIDIA AI Data Platform reference design. It brings AI processing closer to enterprise data while addressing common challenges related to data preparation, governance, and scalability. The release expands the company’s broader strategy of delivering AI-ready data infrastructure for enterprise environments.

As organizations move from AI experimentation to production deployments, many face obstacles related to ingesting and preparing enterprise data, enforcing security and governance policies, and scaling infrastructure to support growing AI workloads. Everpure says Data Stream reduces data preparation timelines from months to minutes while maintaining stream-level access controls that keep data within enterprise boundaries. Its scale-out architecture also allows storage and compute resources to scale independently as AI requirements evolve.

According to Everpure CTO Robert Lee, organizations building AI platforms require flexible architectures that can support both rapid deployment and long-term scaling. He noted that enterprises need secure, high-performance data pipelines that accelerate data processing and reduce time-to-results.

Connecting Data Readiness to Production AI

Everpure positions Data Stream as part of a broader end-to-end AI data platform focused on preparing enterprise information for AI use. The company argues that AI-ready data requires classification, contextualization, governance, security, and scalable access before it can be effectively used for training, inference, or agentic AI applications.

A key component of this strategy is Everpure Data Intelligence, formerly known as 1touch. The platform discovers, classifies, and contextualizes enterprise data across SaaS applications, cloud services, on-premises infrastructure, and mainframe environments. It maps relationships between datasets into a data relationship graph, creating a metadata layer accessible via APIs and the Model Context Protocol (MCP).

The platform also applies attribute-based access controls and governance policies, enabling enterprises to maintain security and compliance requirements as AI models and agents interact directly with business data.

GPU-Accelerated Data Processing

Data Stream is built on the NVIDIA AI Data Platform reference architecture and is designed to simplify the conversion of unstructured enterprise data into AI-ready information. Rather than relying on manual ingestion and data preparation processes, the platform uses a GPU-accelerated pipeline spanning data ingestion through inference.

The goal is to reduce operational complexity while improving the speed at which organizations can deploy AI services and generate actionable results.

NVIDIA Vice President of Storage Technology Jason Hardy said modern AI infrastructure requires architectures that connect secure, governed enterprise data with accelerated computing resources. He noted that Everpure’s integration with the NVIDIA AI Data Platform is intended to help organizations move AI initiatives from proof-of-concept stages into production deployments.

Everpure also disclosed ongoing work on next-generation AI-native storage technologies based on NVIDIA Vera and the NVIDIA BlueField-4 STX storage processor. The effort is focused on bringing acceleration, security, and intelligent data services closer to enterprise datasets as agentic AI deployments continue to expand.

Scaling AI Infrastructure

To address storage bottlenecks that can limit AI training and inference performance, Everpure highlighted FlashBlade as the storage foundation for Data Stream deployments. The platform delivers low-latency data access and incorporates KV Cache Accelerator technology to improve memory efficiency during inference workloads.

Everpure’s Evergreen architecture allows organizations to scale from FlashBlade//S systems to FlashBlade//EXA deployments without disruptive migrations, supporting growth from smaller AI projects to large-scale AI factory environments. Portworx provides the container platform layer for deploying and managing AI pipelines across edge, core, and data center environments.

By combining data intelligence, data streaming, storage, and container orchestration within a unified architecture, Everpure aims to reduce infrastructure fragmentation and eliminate the need for separate AI data silos.

The announcement aligns with findings from a recent IDC Global AI Readiness Survey commissioned by Everpure, which reported that 94% of IT leaders view data quality as the primary factor influencing AI success. Everpure positions its integrated approach as a way for enterprises to maintain flexibility while adapting to rapidly changing AI requirements.

The post Everpure Launches Data Stream to Accelerate Enterprise AI Data Pipelines appeared first on StorageReview.com.