The HP EliteBook X G2i is HP’s premium 14-inch business notebook, and the company is not shy about the AI branding: the formal model name is the EliteBook X G2i 14-inch Notebook Next Gen AI PC. Our review unit runs Intel’s Core Ultra 7 366H, a Series 3 Panther Lake part with 16 cores and vPro, with 32GB of LPDDR5x at 8533 MT/s, a 1 TB SK hynix Gen5 SSD, and a 14-inch 2880 x 1800 display. It is a Copilot+ PC aimed squarely at the corporate fleet, and in our testing, it posted the strongest PCMark 10 Productivity score we have recorded from this class of machine.

HP positions this generation around choice in two places most business laptops treat as fixed. There are two display options, and the shell comes in three finishes: Eclipse Gray, Glacier Silver, and the Atmospheric Blue on our review unit, a splash of personality that is unusual in a segment that defaults to silver and shades of black. Our review unit had two display options: a 1920×1200 OLED with 300 nits of brightness and a 3 K 120 Hz OLED with 500 nits of brightness. Higher resolution and brightness do have the trade-off of increased battery drain, which is quite evident in our battery testing.

Two configuration notes matter. First, HP sent us both members of the family: this EliteBook X G2i and its convertible twin, the EliteBook X Flip G2i. The full benchmark suite reflects the clamshell; we ran battery testing only on the Flip because the two differ meaningfully in display, and we wanted to isolate the panel’s impact on runtime. The Flip earns a brief word in Build and Design but is otherwise not the subject of this review.

Second, and worth being direct about: our unit is not where this platform tops out. HP builds the EliteBook X G2i in two graphics tiers, the 4 Xe-core integrated graphics in our Core Ultra 7 366H unit, which HP positions as the mainstream configuration for everyday productivity, and a 12 Xe-core Arc B390 tier on the Core Ultra X7 358H, paired with a 3K OLED panel, that HP describes as delivering near-discrete GPU performance for analytics, content creation, and local AI work. HP’s own workshop figures claim 3.2 times faster AI image generation and 2.3 times faster hardware-accelerated workflows for the 12 Xe tier, and our comparison data supports the scale of that claim: the same Arc B390 silicon in the Dell Pro 5 14 Intel runs 2.2 to 2.3 times ahead of our review unit in Geekbench GPU compute, Blender, and Stable Diffusion. Every GPU result in this review should be read with that ceiling in mind; we tested the volume build, not the fastest one.

HP does not have a full configurator live for the EliteBook X G2i, so pricing is a snapshot of current stock rather than a build-to-order quote. Our exact build, the Core Ultra 7 366H with the non-touch 3K OLED, was not orderable at this writing. The closest in-stock clamshell is the Arc B390 tier we reference throughout this review: the Core Ultra X7 358H with the same 32GB, 1TB, and 3K OLED fit-out, listed at $6,270 but selling for $3,399 during HP’s current sale. The Flip variant of our 366H configuration sells for $2,999. HP list prices carry heavy standing discounts, so the street figures are the realistic reference points.

HP EliteBook X G2i Specifications

Specification	HP EliteBook X G2i
Model	HP EliteBook X G2i 14-inch Notebook Next Gen AI PC
Processor	Intel Core Ultra 7 366H vPro (Series 3), 16 cores / 16 threads, up to 4.8GHz, 50 TOPS Intel AI Boost NPU
Graphics	Integrated Intel Graphics, 4 Xe cores (HP offers a 12 Xe Arc B390 tier on the X7 358H config)
Memory	32GB LPDDR5x, 8533 MT/s, soldered
Storage	1TB SK hynix PCB01 (HFS001TFM9X187N), PCIe Gen5 NVMe
Display	14-inch 3K (2880 x 1800) OLED, 120Hz, anti-glare, 500 nits, 100% DCI-P3
Operating System	Windows 11 Pro, Copilot+ PC
Battery	68 Watt-hour battery
Ports	1 HDMI 2.1 1 stereo headphone/microphone combo jack 1 nano security lock slot 1 USB Type-A 5Gbps signaling rate (powered) 2 Thunderbolt 4 with USB Type-C 40Gbps signaling rate (USB Power Delivery 3.0, DisplayPort 2.1) 1 USB Type-C 10Gbps signaling rate (USB Power Delivery 3.0, DisplayPort 1.4)
Wireless	Intel Wi-Fi 7 BE211 2×2, Bluetooth 6
Weight	Starting at 0.99kg (2.18 lb) Tested Clamshell 2.4lb/1.088kg, 2-in-1 3.18lb/1.37kg
Colors	Eclipse Gray, Glacier Silver, Atmospheric Blue (as tested)
Price	Exact configuration not currently orderable; closest in-stock build (Core Ultra X7 358H / Arc B390, 32GB, 1TB, 3K OLED) $3,399 at this writing

Build and Design

The EliteBook X G2i is HP’s lead design for its business line, and our review unit makes a strong case for Atmospheric Blue, one of three finishes offered alongside Eclipse Gray and Glacier Silver. This range of color choices is rare in a segment that usually defaults to silver, and in person, the Atmospheric Blue finish feels corporate-professional rather than consumer-focused, with a deep matte tone that hides fingerprints well. The chassis starts at 0.99kg for the lightest build and carries MIL-STD 810H testing, ENERGY STAR certification, and EPEAT Climate+ Gold registration. Our clamshell model weighed 2.4 pounds, and the 2-in-1 model weighed 3.18 pounds, which we believe is mainly due to the aluminum top shell on the 2-in-1 compared to the plastic top shell on the clamshell version.

HP sent us both members of this family, and the pair photographed well together. We benchmarked the clamshell. The EliteBook X Flip G2i convertible folds into tent mode and then into a tablet, and it accepts an optional rechargeable pen that nests inside the chassis. The Flip carries touch panels across its range and gives up the clamshell’s brightest options, which is why it appears in our battery table and nowhere else in this review; the four extra hours it ran are due to the display difference doing the work.

Our unit’s panel is the 14-inch 3K at 2880 x 1800 with a 120Hz refresh, the resolution HP reserves for the top of the display stack. The two-screen strategy is the interesting choice here: the volume configurations run WUXGA at 1920 x 1200 in brightness tiers up to an 800-nit Sure View privacy option, while the 3K tier steps up to OLED, including a 700-nit Tandem OLED touch variant. That is a genuinely wide range for one chassis, and it means two EliteBook X G2i units on the same desk can carry very different screens.

HP pitches the new Tandem OLED at the top of that stack as significantly brighter, with double the panel lifespan and lower power draw. Though our brighter model lacks the all-glass display of its touchscreen counterpart, it still delivers a strong image, backed by its 3k resolution, as shown below. This display would likely perform better in high light or outdoor environments.

The keyboard is HP’s spill-resistant, backlit Premium layout, and its party trick is invisible: the assembly is top-mounted and comes off in three steps with no screws, a swap HP says takes about ten minutes, against competitor designs the company says need many more screws and a full teardown to reach a bottom-mounted keyboard. HP sells a standalone keyboard replacement kit, around $100 to $150 per the workshop, and a QR code engraved on the bottom cover will link straight to replacement ordering. The power key integrates the fingerprint reader, so one press boots and authenticates.

One item people tend not to pay much attention to when purchasing is the trackpad. Here, the trackpad features a haptic design that HP says is about 23% larger this generation, with a waterfall edge that doubles as a slider for brightness and volume, and it is its own module, replaceable without disturbing the keyboard. Our review units have haptic touchpads that let you adjust click sensitivity and haptic intensity directly in Windows settings, along with customizable multi-finger gestures for media, app, and desktop controls.

Above the display, the 5MP IR camera handles Windows Hello and feeds HP’s Poly Camera Pro stack, which runs framing, spotlight, background blur, and virtual backgrounds on the NPU rather than the CPU, part of the AI PC story that holds up in our Procyon NPU results. Dynamic Voice Leveling and AI noise reduction ride along on the audio side.

Connectivity is stronger than the thin chassis suggests. One edge carries HDMI 2.1, two Thunderbolt 4 USB-C ports, and the headset jack; the other adds a third USB-C, a powered USB-A for legacy peripherals, and the lock slot, with an optional nano SIM position for WWAN configurations. Wireless is Intel’s Wi-Fi 7 BE211 with Bluetooth 6.

It is necessary to note that during hands-on testing of the clamshell variant of the EliteBook X G2i, we observed noticeable deflection in the center of the top case. Though it was noticeable when carrying the notebook one-handed and closed, there was not much flex when opening the lid from a single corner. We did not observe any deflection on the 2-in-1 model, but it appears to be due to material differences between the variants. The clamshell model offers lightweight portability, while the 2-in-1 has a bit more heft in exchange for a stiffer top case for the flip function.

The underside is a single vent field feeding two fans. HP’s SmartSense uses motion sensors to distinguish a lap from a table and cools the underside accordingly. The Smart Resource Optimizer prioritizes foreground applications, delivering up to 15% better performance under heavy load, according to HP. A Battery Extender mode on the Intel models trades refresh rate, brightness, and conferencing effects for what HP says adds up to two extra hours.

Inside, the layout is as serviceable as the keyboard story implies: two fans up top, the 68Wh Long Life 6-cell pack across the bottom, and the M.2 SSD accessible under a shield. The battery fast-charges to 50% in 30 minutes, and HP also ships a worldwide battery replacement kit for end-to-end swaps. For an IT fleet, this, plus the replaceable top-mount keyboard, is the difference between a bench repair and a chassis swap. The sustainability story runs deep: HP says 80% of the major parts contain recycled materials, down to magnesium from recycled wheel rims and carbon black from tires, and the redesigned 100W GaN adapter in our box is at least 90% recycled plastic.

HP EliteBook X G2i Performance

Our review unit runs the Core Ultra 7 366H with integrated Intel graphics, 32GB of LPDDR5X at 8533 MT/s, and a 1 TB SK hynix Gen5 SSD on Windows 11 Pro, with benchmarks tested in the Best Performance power mode. For battery life testing, we configure systems into Balanced power mode and set the screen brightness to 50%.

For comparables, we included the Dell Pro 7 14 Intel (Core Ultra 7 366H, 64GB), which runs the same processor and integrated graphics as our review unit and gives us a direct silicon-to-silicon read; the Dell Pro 5 14 Intel (Core Ultra X7 368H, Arc B390, 64GB), whose 12 Xe-core GPU is a useful proxy for the EliteBook X G2i configuration we did not test; and the Lenovo ThinkPad P14s Gen 7 (Core Ultra 7 366H, RTX PRO 1000, 64GB) as the workstation-class reference, again on the same CPU. Note that all three comparison systems have 64GB of memory compared to our unit’s 32GB, and a few comp cells are N/A where a system’s review predates a benchmark in the current suite.

PCMark 10

PCMark 10 measures general system performance across everyday work such as web browsing, video conferencing, spreadsheets, and content creation, providing a broad view of where a system’s strengths lie. Higher scores are better.

PCMark 10	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Overall Score	9,209	8,438	7,945	9,083
Essentials	10,760	10,981	10,751	10,686
Productivity	18,859	13,992	13,821	16,501
Digital Content Creation	10,445	10,610	9,158	11,534

 

The overall 9,209 leads every comparison system, but the Productivity sub-score is the headline: 18,859 is the highest we have recorded in this class, 14% ahead of the ThinkPad and roughly 35% ahead of both Dell Pro machines, driven by a 23,502 spreadsheet score. Digital Content Creation lands mid-pack, which fits an 8 Xe-core iGPU against the Arc and RTX silicon in the group.

PCMark 10 Modern Office Battery

The battery test repeatedly runs common office tasks until the battery reaches the test’s cutoff point, in Balanced mode at 50% display brightness. This is a rundown of the whole system rather than a synthetic idle drain, so it tracks closely with what a full day of productivity work does to the battery. Longer runtimes are better. This table adds the EliteBook X Flip G2i, which we battery-tested specifically to isolate the display’s impact on runtime; the two systems share a platform but carry different panels.

Modern Office Battery	HP EliteBook X G2i	HP EliteBook X Flip G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Runtime (higher is better)	17 hours 13 minutes	21 hours 20 minutes*	26 hours 18 minutes	26 hours 48 minutes	15 hours 52 minutes

 

*The Flip result is the runtime recorded in the test log; the run ended with a benchmark error after logging 21 hours and 20 minutes.

Seventeen hours and 13 minutes clear two working days, but it sits mid-pack here: both Dell Pro machines run past 26 hours, and the Flip outlasts its own clamshell sibling by more than four hours on the same silicon. That gap is the display talking, and it is the clearest evidence in this review that panel choice is the battery decision on this platform. HP’s own workshop testing tells the same story at a larger scale: the company clocked around 31 hours of video playback on the 300-nit OLED against around 23 hours on the 500-nit panel, per HP’s internal figures.

Geekbench 6

Geekbench 6 measures processor performance using a mix of common tasks, with separate scores for single-core and multi-core workloads, plus GPU compute scores through OpenCL and Vulkan. Higher scores are better. The Dell Pro reviews predate our GPU compute captures on this test, so those cells are N/A.

Geekbench 6	HP EliteBook X G2i (Intel Graphics)	Dell Pro 7 14 Intel (Intel Graphics)	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
CPU Single-Core	2,804	2,862	2,968	2,808
CPU Multi-Core	16,390	16,787	16,874	16,319
GPU OpenCL	24,453	23,741	32,458	87,537
GPU Vulkan	26,528	25,533	43,382	73,204

 

The three 366H systems land within 3% of each other on CPU, with the X7 368H in the Dell Pro 5 14 taking the single-core lead. The review unit’s integrated graphics turn in 24,453 in OpenCL and 26,528 in Vulkan, numbers that the RTX PRO 1000 more than triples; within integrated silicon, they are exactly where this iGPU should sit.

Geekbench 7

Geekbench 7 joins the suite alongside Geekbench 6 as comparison data accumulates. It uses updated workloads and a rebased scoring range, so results are not comparable between the two versions. Higher scores are better.

Geekbench 7	HP EliteBook X G2i (Intel Graphics)	Dell Pro 7 14 Intel (Intel Graphics)	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
CPU Single-Core	2,527	2,589	2,701	2,533
CPU Multi-Core	17,345	17,805	18,737	17,651
GPU OpenCL	23,737	23,876	54,664	75,340
GPU Vulkan	23,036	23,484	47,233	70,067

 

The silicon-to-silicon read is tight: 2,527 and 17,345 against the Dell Pro 7 14’s 2,589 and 17,805 on the same processor, and the two integrated GPUs land within 2% of each other in both APIs. The Arc B390 in the Dell Pro 5 14 more than doubles both iGPU compute scores, which is the cleanest preview of what the untested X7 358H EliteBook config would add.

Cinebench 2026

Cinebench 2026 is the current release in the Cinebench line and the only version we report. It tests CPU and GPU performance using Maxon’s Redshift render engine, built on the latest Cinema 4D code. Because code and compiler changes accelerated scene rendering, scores use an adjusted range and should not be compared to previous Cinebench versions. Its GPU test does not run on Intel integrated graphics, so only the discrete-GPU ThinkPad posts a GPU score here.

Cinebench 2026	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
CPU Single Thread	494	511	530	502
CPU Multiple Threads	4,060	3,807	3,656	4,492
GPU	N/A	N/A	N/A	34,437

 

The review unit posts the best multi-thread result of the three integrated systems at 4,060, 7% ahead of the same-CPU Dell Pro 7 14, though the RTX-cooled ThinkPad still leads at 4,492. Single-thread runs the other way, with the EliteBook at the back of a pack separated by 7%.

3DMark CPU Profile

3DMark CPU Profile measures CPU performance at fixed thread counts, showing how a system scales from single-threaded work up to full saturation. Higher scores are better. The Dell Pro reviews predate this test in our suite, so their cells are N/A.

3DMark CPU Profile	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Max Threads	9,179	9,699	10,370	10,500
16 Threads	8,995	9,679	10,355	10,421
8 Threads	5,814	6,450	6,669	6,550
4 Threads	3,812	4,240	4,256	4,219
2 Threads	2,095	2,230	2,269	2,251
1 Thread	1,108	1,163	1,181	1,165

 

Scaling is orderly from 1,108 at one thread to 9,179 at max threads, but the ThinkPad leads at every step on the same processor, by 13% at full saturation. Sustained-load thermal headroom is the likely difference between a 3-pound business chassis and a thicker workstation build.

7-Zip Compression

The built-in 7-Zip benchmark measures how quickly the processor can compress and decompress data using multiple threads, run with a 128MB dictionary across ten passes. Higher GIPS scores are better. The Dell Pro reviews recorded only the total rating, so their compression and decompression cells are N/A.

7-Zip 24.09 (GIPS)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Compressing	89.259	83.281	93.417	90.364
Decompressing	86.189	76.774	89.181	90.526
Total Rating	87.724	80.027	91.299	90.445

 

A total rating of 87.724 GIPS beats the same-CPU Dell Pro 7 14 by 8% and lands within 3% of the group-leading ThinkPad. For a 32GB system against 64GB comps, memory capacity clearly is not the constraint in this workload.

y-cruncher

y-cruncher measures how quickly the processor can calculate a large number of digits of Pi, placing a heavy load on the CPU and memory subsystem, while the BBP extracts hexadecimal digits of Pi. Results are in seconds, so lower times are better. The review unit could not run the 5-billion-digit computation because it exceeded the 32GB of memory, but in the lower tests, it held its own against the Dell units.

y-cruncher (seconds, lower is better)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Pi 1B	34.920	34.775	29.405	26.685
Pi 2.5B	97.527	104.586	90.719	76.787
Pi 5B	N/A	240.554	204.685	172.994
Pi BBP 1B	1.842	1.835	1.718	1.621
Pi BBP 10B	22.451	21.280	20.212	18.200
Pi BBP 100B	269.011	301.611	256.416	219.969

 

At 1 billion digits, the review unit and the Dell Pro 7 14 are separated by a rounding error, which is what identical silicon should do. The 5B row is the one to notice: this is the first concrete cost of the 32GB configuration in this review, a workload the 64GB comps simply complete, and this machine cannot start.

Blender

The Blender benchmark measures rendering performance using three different 3D scenes: Monster, Junkshop, and Classroom. Results are reported in samples per minute, so higher scores are better. We test on the CPU and GPU. Scores are not comparable across Blender versions, so we report the scores for the current 5.2 release here.

Blender 5.2 (samples/min)	HP EliteBook X G2i (Intel Graphics)	Dell Pro 7 14 Intel (Intel Graphics)	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
GPU
Monster	250.14	254.40	560.28	922.95
Junkshop	154.95	189.59	466.36	795.65
Classroom	149.90	152.73	414.57	625.62
CPU
Monster	118.87	107.87	129.61	131.08
Junkshop	84.95	78.06	90.52	97.76
Classroom	57.45	53.90	62.20	68.45

 

GPU rendering tracks the silicon exactly: 250 samples per minute on Monster against 254 for the same iGPU in the Dell Pro 7 14, while the Arc B390 in the Dell Pro 5 14 renders at 2.2 times that rate and the RTX PRO 1000 at 3.7 times. CPU rendering is more flattering, with the review unit ahead of its silicon twin on all three scenes.

LuxMark

LuxMark is an OpenCL rendering benchmark built on LuxCoreRender, run on the Hall and Food scenes. Higher scores are better.

LuxMark v4	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Hall	2,481	2,168	3,287	11,342
Food	1,058	880	1,585	4,103

 

Hall at 2,481 and Food at 1,058 put the review unit 14% and 20% ahead of the same-CPU Dell Pro 7 14, with the usual integrated-versus-discrete gulf beyond that: the Arc B390 adds roughly a third again, and the ThinkPad’s RTX PRO 1000 is in a different category entirely.

V-Ray

V-Ray measures GPU path-tracing performance in vpaths using Chaos’ production renderer. Higher is better. The RTX engine requires NVIDIA hardware, so only the ThinkPad posts a score there.

V-Ray GPU (vpaths)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
CUDA Engine	850	775	919	1,568
RTX Engine	N/A	N/A	N/A	2,589

 

The 850 vpaths through the CUDA-compatible path are 10% up on the silicon twin and 8% behind the Arc B390. As with every GPU render test in this review, the class jump is from the discrete card to the integrated tier, not the other way around.

Blackmagic RAW Speed Test

The Blackmagic RAW Speed Test measures how quickly a system can decode Blackmagic RAW footage on the CPU and GPU, reported in frames per second at 8K with 12:1 compression. Higher is better. The Dell Pro reviews predate this test in our suite, so their cells are N/A.

Blackmagic RAW Speed Test	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
8K 12:1 CPU (fps)	72	68	76	77
8K 12:1 GPU (fps)	48	48	86	96

 

CPU decode at 72fps sits just behind the ThinkPad’s 77, while the GPU path at 48fps is half the RTX PRO 1000’s rate. For proxy-free 8K editing, this is a machine that leans on its CPU.

Topaz Video AI

The Topaz Video AI benchmark measures AI video upscaling and frame-interpolation performance in frames per second across the application’s enhancement models, run at 1080p input, where higher is better. The 16X Slowmo Aion model failed to complete on the review unit, the same failure we have recorded on other Intel integrated graphics systems.

Topaz Video AI (fps)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
Artemis 1X / 2X / 4X	2.45 / 2.48 / 0.91	2.51 / 2.53 / 0.95	6.27 / 5.23 / 1.88	5.17 / 3.19 / 1.12
Iris 1X / 2X / 4X	2.44 / 1.40 / 0.43	2.38 / 1.37 / 0.42	5.06 / 3.10 / 0.93	5.91 / 3.12 / 1.01
Proteus 1X / 2X / 4X	2.59 / 2.86 / 1.11	2.52 / 2.76 / 1.06	6.37 / 6.03 / 2.42	4.78 / 3.14 / 1.05
Gaia 1X / 2X / 4X	1.34 / 0.91 / 0.69	1.31 / 0.89 / 0.67	3.27 / 2.24 / 1.47	1.62 / 1.13 / 0.79
Nyx 1X / 2X	0.72 / 0.72	0.71 / 0.71	1.56 / 1.52	2.40 / 2.09
Hyperion HDR 1X	2.08	2.02	3.13	14.56
4X Slowmo Apollo / APFast	4.61 / 16.06	3.08 / 14.64	8.52 / 24.27	10.39 / 29.94
16X Slowmo Aion	DNF	DNF	DNF	N/A

 

The review unit and the Dell Pro 7 14 move in lockstep across every model, usually within a few hundredths of a frame, which is silicon parity on display. The Arc B390 roughly doubles throughput on the enhancement models, and the Aion interpolation failure remains an Intel iGPU pattern rather than an HP problem.

UL Procyon AI Text Generation

The Procyon AI Text Generation benchmark measures local LLM inference across four models of increasing size—Phi, Mistral, Llama3, and Llama2—run here using OpenVINO on the integrated GPU. The Pro 5 14 does bring a higher-class CPU than the other comparisons, as well as a higher-class iGPU, but we will have to see how it stands against the RTX PRO 1000 that the ThinkPad P14s Gen 7 brings. When testing the RTX PRO 1000, we have to use a different inference engine: ONNX Runtime, as OpenVINO is Intel-only.  Higher scores are better.

Procyon AI Text Generation	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7 (IGPU / RTX PRO 1000)
Phi	1,082	1027	1772	1071 / 1,618
Mistral	1,053	1027	1767	1058 / 1,397
Llama3	952	1014	1741	1047 / 1,252
Llama2	993	1009	1731	1012 / DNF

 

We see a pretty tight comparison with the same silicon in the Dell Pro 7 14, with a 55-point gap on Phi and a 26-point gap on Mistral. Llama 3 widens the gap in the other direction, favoring Dell with a 62-point lead and a 16-point lead over Llama 2. The Pro 5 14 leads the comparison group by a wide margin, helped by its higher-class CPU and iGPU, and even outpaces the RTX PRO 1000 in this test. That result likely reflects the Arc B390’s unified memory setup compared with the dedicated VRAM available to the RTX GPU.

UL Procyon AI Image Generation

The Procyon AI Image Generation benchmark measures local Stable Diffusion inference, run here through OpenVINO on the integrated GPU. The INT8 test uses the quantized SD 1.5 model. Higher scores are better.

Procyon AI Image Generation	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
SD 1.5 FP16	258	258	635	943
SD 1.5 INT8	3,529	3,575	7,693	12,403
SDXL FP16	307	268	646	765

 

SD 1.5 FP16 returns 258, matching the Dell Pro 7 14 to the point, and INT8 lands within 1.3%. Just like above, the Dell Pro 5 14 brings a bit of a performance boost with its higher-class iGPU, and the ThinkPad P14s Gen 7 shows off its RTX PRO 1000.

UL Procyon AI Computer Vision

The Procyon AI Computer Vision benchmark measures inference performance across a range of neural networks, including MobileNet V3, ResNet 50, Inception V4, YOLO V3, DeepLab V3, and REAL-ESRGAN. The WinML runs use float32 on CPU and GPU, giving a like-for-like view across vendors. Higher scores are better.

Procyon AI Computer Vision (WinML)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
CPU	126	121	119	134
GPU	205	205	398	426

 

In the computer vision test, we see tight comparisons again with 126 on CPU and 205 on GPU, almost matching the Dell Pro 7 14’s 121 on CPU and 205 on GPU scoring. It still comes in behind the Dell Pro 5 14 and the Lenovo, which have better GPUs, but the CPU scores were all within 15 points of each other.

SPECviewperf 15

SPECviewperf 15 measures graphics performance using viewsets derived from professional applications in CAD, 3D modeling, rendering, engineering, and medical visualization, replayed here at 1080p. Higher scores are better, although this is a business notebook rather than a certified workstation, so these results are context rather than a purchase criterion.

SPECviewperf 15 (FHD)	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel (Arc B390)	Lenovo ThinkPad P14s Gen 7 (RTX PRO 1000)
3dsmax-08	8.70	9.58	20.21	28.20
blender-01	9.26	9.11	21.19	40.69
catia-07	5.66	5.27	10.27	43.36
creo-04	18.85	18.27	32.53	107.59
energy-04	3.62	3.68	10.78	45.79
enscape-01	6.29	6.22	14.28	25.45
maya-07	52.56	49.54	82.42	112.84
medical-04	9.97	9.91	22.96	86.38
snx-05	37.45	37.74	46.22	103.61
solidworks-08	12.37	11.86	23.60	53.63
unreal_engine-01	21.47	21.36	38.77	49.24

 

The review unit and the Dell Pro 7 14 are within a few percent across all 11 viewsets, another clean silicon match. The Arc B390 roughly doubles most viewsets, and the RTX PRO 1000 runs far ahead, which is the certified-workstation premium in one table.

SPECworkstation 4

SPECworkstation 4 measures workstation performance across CPU, graphics, storage, AI, and industry-vertical workloads using real applications. Higher scores are better, and N/A means the system did not complete all workloads required for that category, while DNF indicates a failed workload.

SPECworkstation 4	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
Hardware Subsystems
CPU	1.16	1.08	1.19	1.35
Graphics	0.80	0.82	1.68	4.51
Accelerator	1.46	N/A	N/A	3.26
Storage	1.65	1.68	1.76	1.67
Industry Verticals
AI & Machine Learning	1.16	1.18	1.36	1.65
Energy	1.02	0.90	1.18	1.69
Financial Services	0.79	0.78	0.78	0.96
Life Sciences	1.14	1.04	1.34	1.82
Media & Entertainment	1.24	1.16	DNF	1.81
Product Design	1.47	1.41	1.64	1.89
Productivity & Development	0.93	1.04	1.10	1.34

 

The review unit completed all 23 workloads cleanly, including the 7-Zip workload that has recently tripped several systems in our lab. The scores themselves tell the business-laptop story: competitive CPU and storage subsystems, a 0.80 graphics subsystem reflecting the 8 Xe cores, and an Accelerator score of 1.46 from the NPU that neither Dell Pro nor the original run captured.

Storage Performance

We measure storage with 3DMark’s Storage benchmark, which replays real-world gaming and content traces, and the Blackmagic Disk Speed Test for sequential throughput. Higher is better for both.

Storage	HP EliteBook X G2i	Dell Pro 7 14 Intel	Dell Pro 5 14 Intel	Lenovo ThinkPad P14s Gen 7
3DMark Storage	3,156	3,259	3,144	3,094
Blackmagic Write (MB/s)	8,949.2	8,934.5	8,747.3	8,262.3
Blackmagic Read (MB/s)	7,919.3	8,398.6	8,609.6	8,511.5

 

The SK hynix Gen5 drive writes at 8,949 MB/s, the fastest in the group, and its 3DMark Storage score of 3,156 sits in the middle of a tightly packed field. Sequential read at 7,919 MB/s trails the comps by 6 to 8%, a gap nobody outside a synthetic benchmark will feel.

Conclusion

Through performance testing, the EliteBook X G2i held its own against the Dell Pro 7 14 on the same silicon, ahead in Cinebench 2026 multi-thread and LuxMark, and a step behind in Geekbench 7, while coming in meaningfully lighter. Battery life trailed the LCD-equipped Pro 7 14 at 17 hours 13 minutes versus 26:18, but most of that gap is due to the 3K OLED panel’s power draw. The 32GB LPDDR5x configuration cost it a little in memory-bound workloads, where the largest y-cruncher runs were out of reach; a 64GB option is available in the lineup for buyers who need the headroom.

For buyers who want more graphics, HP offers the Core Ultra X7 358H with Arc B390 graphics in this same chassis, a 12 Xe-core tier that should land close to the Dell Pro 5 14 columns in our charts, since both carry the same CPU and GPU package. As it happens, that is also the build HP is actually selling: with no configurator live, the closest in-stock clamshell to our unit is the B390 version with the same 32GB, 1TB, and 3K OLED fit-out at $3,399 on sale, while the Flip variant of our 366H configuration runs $2,999.

Our one real gripe on the build is lid flex. Pressing the center of the closed lid produces noticeable give that competitors, and even the Flip counterpart, resist better, though opening the lid from a single corner shows little deflection. Beyond that, the chassis feels premium, and the machine is comfortable to work on.

The EliteBook X G2i runs head-to-head with heavier competition in a 2.4-pound package with a respectable port selection, and HP’s repairability work deserves the call-out: the keyboard comes out from the top without screws, and the battery swap is also easy. For buyers who want a serviceable, genuinely light business machine with a panel worth looking at all day— and even three color options—this is a solid choice.

For configuration options and current pricing, visit the HP EliteBook X G2i product page.

Leaderboard: The HP EliteBook X G2i enters our Laptop Battery Life Leaderboard at #9 with 17 hr 13 min, its Flip sibling enters at #7 with 21 hr 20 min, and it takes Best Executive Thin-and-Light on the Best Business Laptops leaderboard.

The post HP EliteBook X G2i Review: 17 Hours of OLED Battery in a 2.4-Pound Business Flagship appeared first on StorageReview.com.

The Enterprise NAS (ENAS) is Ubiquiti’s first run at more enterprise-lite business storage. The 3U chassis holds 16 hot-swap drive bays, an eight-core Arm Neoverse N2 processor, 64GB of ECC memory, dual 25GbE SFP28 ports, and redundant hot-swap 550W power supplies, all for $3,999 MSRP. One pricing note: Ubiquiti’s store currently applies a memory surcharge at checkout, its response to the memory price environment, so out-the-door pricing runs above list until component costs settle. The filesystem underneath is ZFS, the drive bays accept whatever disks you want to put in them, and every software feature ships with the box. There are no licenses, no feature unlocks, and no mandatory support contracts.

The price for a bare unit might look daunting at first, but there is a lot of value to be found. Dual 25GbE as standard equipment is the piece competitors do not match: Synology’s 16-bay RS4021xs+ ships with two 10GbE RJ45 ports and puts 25GbE behind a PCIe add-in card, and the story is much the same across rackmounts at this capacity point, where faster-than-10GbE networking is another line on the quote. Pair that with redundant power and ECC memory, and Ubiquiti is checking boxes that have traditionally separated SMB NAS boxes from entry-level enterprise arrays. ENAS is available now directly from Ubiquiti (affiliate link).

ENAS also caps a portfolio expansion that has been running at a remarkable pace. Ubiquiti built its name on access points and gateways, and the networking side keeps rolling; in the past year alone, we have looked at the UniFi E7 and E7 Campus WiFi 7 access points, the Cloud Gateway Fiber, and the Dream Router 7. Storage is newer ground. The line started with the UNAS Pro, a $499 seven-bay 2U unit that made sense mostly for UniFi loyalists, then grew with the compact UNAS 2 and the UNAS Pro 8, which added redundant power and NVMe caching. ENAS is the first of the family with ZFS, ECC memory, native iSCSI, and SAS expansion ports, and it is worth noting that the file systems differ enough that Ubiquiti says UNAS data cannot be restored directly to an ENAS; only users, groups, and settings carry over.

The segment ENAS enters is not empty. Synology and QNAP, along with others, have owned the SMB NAS market for two decades, and their platforms remain far deeper in software: app ecosystems, containers, surveillance suites, and mature backup tooling that UniFi Drive does not attempt to match today. But the incumbents have provided challengers with an opening. Synology spent most of 2025 enforcing a compatibility policy that effectively locked its 2025 Plus-series units to validated, largely Synology-branded drives, before walking the policy back with DSM 7.3 in the fall. Against that backdrop, a 16-bay ZFS system with an explicit open-drive stance and no software licensing is a well-aimed product.

The buyer this makes the most sense for is easy to picture: a small or mid-sized shop, or the MSP that runs it, with a moderate-to-large UniFi estate already in place. For those environments, ENAS drops into the same console as the switches, gateways, access points, and cameras, inherits Site Manager for multi-site visibility, and ties into the same identity model, with UniFi Endpoint handling remote file access. Storage stops being a separate vendor with a separate UI and a separate renewal. For shops that need easy, reliable file and block storage and already understand the UniFi interface, operational continuity is as much the product as the hardware.

One note on our test unit: we have had ENAS in the lab since November 2025, and the production hardware revision moved the rear I/O from 10GbE to the dual 25GbE configuration shipping today. Everything pictured and tested here reflects the current retail hardware, and the software has been moving quickly since our unit arrived, a point we will come back to.

Ubiquiti Enterprise NAS Specifications

Specification	Ubiquiti Enterprise NAS
Overview
Dimensions	481.4 × 480 × 132 mm (19 × 18.9 × 5.2 in)
Storage Capacity	16 × 2.5/3.5-inch drive bays 2 × M.2 NVMe bays
Networking Interface	2 × 25G SFP28 (25G/10G/1G) 1 × 10GbE RJ45 (10G/5G/2.5G/1G/100M)
Expansion Port	2 × SFF-8644 (24G)
Power Redundancy	Supported
Form Factor	3U rackmount
Hardware
Drive Support	16 × 2.5/3.5-inch HDD/SSD 2 × M.2 NVMe SSD 2 × Expansion ports
Max Power Budget for Drives	450W
Max Power Consumption	550W
Power Method	Dual AC input, hot-swappable power modules
Power Supply	2 × Hot-swappable AC/DC 550W power modules
Processor	Eight-Core ARM N2 at 2.4GHz
Memory	64GB
Management	Ethernet
RF Interface	Bluetooth 4.1
Weight	16.1 kg (35.5 lb)
Enclosure Material	SGCC steel
Mount Material	SGCC steel
Supported Rack Depth	Rails support 600 mm (23.6 in) four-post racks with square holes (9.5 × 9.5 mm) Post depths from 600–1066 mm (23.6–42 in)
Faceplate	3U Bezel (4.7-inch touchscreen, RGBW LEDs) 3U Bezel Lite (blank) Both optional
LEDs
Status LEDs	Ethernet, SFP28, HDD, System, Expansion Port, CRPS
Environmental & Compliance
Operating Temperature	-5°C to 40°C (23°F to 104°F)
Operating Humidity	5% to 95% noncondensing
NDAA Compliant	Yes
Certifications	FCC, CE, IC
Software
Supported File Protocols	NFS, SMB
Supported Block Protocols	iSCSI
RAID Types	Mirror, RAID-Z1, RAID-Z2, RAID-Z3
RAID Groups	Multiple
Hot Spare Support	Supported
Personal & Shared Drives	Supported
SSD Cache	Supported
Max NVMe SSD Capacity	8 TiB
File Encryption	Supported
Backup Support	Remote UNAS, CIFS/SMB server, cloud services
Cloud Backup Services	Google Drive, OneDrive, Dropbox, Amazon S3, Backblaze B2, Wasabi
Snapshots	Supported
Share Links	Supported
Time Machine Backup	Supported
Client App Support	Supported
User Groups	Supported

Build And Design

The ENAS is built to a different standard than the rest of the UniFi storage line, closer to the rackmount servers that pass through the lab than to the UNAS Pro it descends from. The differences start at the drive bays. ENAS caddies use a proper release button with the activity LED set into its center, replacing the push-to-release caddies on the UNAS Pro. It is a small change that makes hot-swaps less fiddly when the rack is dark and a drive needs pulling. A plug at the top right of the front panel covers a connector we will come back to at the end of this section.

The aluminum handles on the rack ears do two jobs: they release the rackmount latches, and they anchor the optional bezel. Ubiquiti ships the ENAS with King Slide rails, the same family we see on servers from the major OEMs, and mounting them into the rack is tool-less. Getting the inner rail onto the chassis is not easy and takes a screwdriver and the screws along the bottom of the unit. Once that is done, the ENAS slides in and locks with the latches at the top of each handle.

The caddies are also where some cost engineering shows. The release lever is plastic, whereas the UNAS Pro used what feels like aluminum, a downgrade on the one part that gets handled every time a drive moves. Mounting is tool-less for 3.5-inch drives, with an optional locking screw for anyone who wants it, but 2.5-inch drives still take four screws through the bottom of the caddy. Fill all 16 bays with SSDs, and that is 64 screws standing between you and a populated chassis.

The rear panel carries an RJ45 console port, twin 25G SFP28 ports, and twin SFF-8644 connectors, better known as external Mini-SAS HD. Each Mini-SAS port carries 24G to a shelf, supporting up to two of the UACC-ESE-16 Enterprise Storage Expansion, a no-frills 16-bay expansion chassis. Ubiquiti specifies the shelf as a 3U unit with a single SFF-8644 port and the same pair of hot-swap 550W power modules used in the ENAS, and lists it as compatible with the ENVR-Core as well. It appeared on Ubiquiti’s site while we were finishing this review but had not reached the store, so we have not had one in the lab. Two shelves take the system to 48 bays, which is where Ubiquiti’s claim of more than a petabyte of raw capacity comes from.

A blank panel to the right of the expansion ports opens onto two M.2 slots. Trays are not included, so populating them means adding Ubiquiti’s UACC-SSD-Tray (Affiliate Link). UniFi Drive uses these as a read-only SSD cache rather than as a general-purpose fast tier. Ubiquiti does offer a read-write cache mode on the UNAS line, but the ZFS-based ENAS does not support it.

Further to the right sit the redundant hot-swap power supplies, a pair of 550W CRPS modules. Nothing about them is novel to anyone who works on servers, which is rather the point: this is the first UniFi storage box whose parts bin looks like the rest of the rack. Each module releases with a locking tab on the side and pulls out on a foldaway handle.

For a $3,999 chassis, the back panel is dense. Console, dual 25G, and SAS expansion on one plate is a spec sheet normally found a tier or two further up, and it is the clearest signal that ENAS is not simply a bigger UNAS.

And finally, the small port mentioned on the front of the unit at the beginning of this section is a magnetic USB Type-C connector for the optional UniFi 3U Bezel, which Ubiquiti specs with a 4.7-inch touchscreen and RGBW status lighting drawing under 4W. If screens and lights are not your thing, but you want to cover your drive bays, Ubiquiti offers the blank UACC-3U-Bezel-Lite. Both bezels are listed on Ubiquiti’s site but have not reached the store as of this writing, so we have not had either one in hand.

That bezel is more than decoration, because the ENAS has no power button. Shutting the system down from the interface is easy; getting it back up is not. With no front-panel control, the only way to power the unit on while standing at the rack is to pull both power cords and reseat them. Remotely, there is no power-on control in the UniFi interface either, which leaves Wake-on-LAN as the only option. For a box destined for closets and small racks that may never see a KVM, that is worth knowing before it ships to a site an hour away.

The build lands where Ubiquiti hardware usually does, clean on the outside with most of the practical details right: tool-less rail mounting, tool-less 3.5-inch caddies, redundant power, and 25G on board. The Mini-SAS ports are the part that changes the math later, since they turn a 16-bay box into a 48-bay one without replacing anything already installed. The missing power button is the one real misstep.

ENAS Management

ENAS is managed through UniFi Drive on UniFi OS, which means anyone who has adopted a UniFi device in the past will find nothing surprising in the workflow. The system appears in the same console as the rest of a UniFi deployment, and Site Manager provides fleet-level visibility across sites so that an MSP can watch capacity and health on a client’s ENAS alongside their switches and access points. Local storage administration covers ZFS pool creation in Mirror, RAID-Z1, RAID-Z2, or RAID-Z3 layouts, shared folders over SMB (including Time Machine targets for macOS clients) and NFS, and iSCSI LUNs for block workloads. Scheduled snapshots, replication between UniFi storage devices, and backup jobs to third-party cloud targets, including Amazon S3, Backblaze B2, Wasabi, Google Drive, OneDrive, and Dropbox, round out the data protection story, and AD/LDAP integration handles user import for shops with existing directories. Drive encryption is supported, as are ZFS inline compression and deduplication, though Ubiquiti sensibly cautions that dedup is resource-hungry and best reserved for SSD-backed pools.

The two external M.2 slots serve as an SSD cache rather than a general storage tier. Ubiquiti offers read-only and read-write cache on the UNAS line, although for the ENAS, the SSD cache is read-only with one or two SSDs. The cache deliberately bypasses large sequential files, since streaming them through the SSDs would burn endurance for no real gain, so the payoff lands in concurrent random access, office file sharing, and database-style workloads rather than large media pulls. We tested with and without the cache populated, and the results below bear that behavior out.

The caveat is that the feature set is still thin next to DSM or QTS. There is no application ecosystem, no container or VM hosting, and no equivalent of Synology’s Active Backup suite; ENAS is a storage appliance, not an app platform yet. The counterweight is how fast the platform is moving. Since our unit arrived, Ubiquiti has shipped a steady cadence of UniFi Drive releases, most recently version 4.3.10 on July 30, and the company has publicly committed to immutable snapshots, multi-site backup orchestration, and expansion shelves that Ubiquiti says will take a single system past a petabyte of raw capacity through the two SFF-8644 ports. Buyers should evaluate ENAS on what it does today, but the trajectory is important to consider, and the gap is closing quarter by quarter.

The landing page in UniFi Drive condenses the entire system into a single view. A status rail on the left covers the chassis: network address, uptime, firmware level (UniFi OS 5.1.19 and Drive 4.3.6 during our time with the unit), CPU temperature, and a three-position fan profile slider. The center pane tracks live disk throughput, while the panel below breaks out the pool with per-share usage and a snapshot schedule column, each share a click away from protection. Our RAID-Z2 pool showed 132.76TB in use of 170.57TB usable, with dedupe and iSCSI carve-outs itemized on the capacity bar.

Storage Pools are where the ZFS decisions get made, and Ubiquiti does more hand-holding here than most. Illustrated cards across the top explain the trade-offs between a single RAID group, multiple RAID groups in one pool, multiple pools with different protection levels, and global hot spares, useful guardrails for an admin who has never touched ZFS. Below that sits our test pool: a single RAID-Z2 group of eight 30TB 7,200RPM hard drives, good for two-drive failure protection, with bays 9 through 16 left open.

Block storage gets its own view. LUN creation is a short wizard, and the list shows host count, map status, per-LUN usage, and the backing pool at a glance. Initiator access is controlled through host groups rather than per-LUN fiddling; we provisioned a pair of 25GB LUNs mapped through a single host group for our iSCSI testing. It is minimal compared to what a dedicated SAN interface exposes, but it covers the Proxmox/VMware/Hyper-V shared-storage case that ENAS is aimed at.

The Services page collects protocol plumbing in one place: the web access endpoint, SMB with copy-ready paths for macOS and Windows, NFS exports with per-host permissions and squash settings alongside a pre-built mount command, plus Time Machine and Rsync toggles. One quirk worth noting is that file services use their own credentials, separate from UniFi accounts; a banner reminds you that they are required before SMB mounts or Time Machine will work.

Day-to-day administration happens in the file view, where shared drives are listed with active user counts, usage, pool assignment, and protection status. This is where our 314T share, holding the bulk of the 125TB test data set, lived alongside a pair of smaller shares. Personal drives are provisioned per user through Admins & Users, which keeps end-user storage separate from departmental shares without extra configuration.

Performance Testing

To exercise the ENAS over its shipping 25GbE interfaces, we built a single RAID-Z2 group from eight Seagate IronWolf Pro 30TB CMR drives in bays 1 through 8, leaving the back half of the chassis empty, and generated load from a Dell PowerEdge R740 running Ubuntu Server with a dual-port NVIDIA ConnectX-4 25GbE NIC, directly attached to the ENAS SFP28 ports. Our testing used FIO sweeps of 4K random and 1M sequential transfers, in both read and write, across a matrix of queue depths and job counts to find each protocol’s peak and its saturation behavior.

We ran the sweep against six configurations: NFS with asynchronous mounts, NFS with synchronous mounts, and iSCSI, each with and without the M.2 cache populated. The cache drives are Ubiquiti-supplied 1TB M.2 units that report as Kingston OM8SEP41024Q-A0, a TLC NVMe part from Kingston’s OEM catalog rather than an enterprise SSD. For a read-only cache that is a sensible fit, since the device serves reads and never has to absorb sustained writes. Charts below plot all six against each other; peak figures are called out in the text.

4K Rand Read IOPS

Random 4K reads cluster tightly across every NFS configuration, peaking between 89.5K and 90.9K IOPS regardless of sync mode or cache. With a working set this size against 64GB of ECC memory, ZFS is serving the bulk of these hits from ARC, so the ceiling reflects the network path and protocol stack more than the spinning disks underneath. iSCSI tops out lower, at 40.4K IOPS (39.7K with the cache populated), a little under half the NFS figure.

4K Rand Read Latency

Latency tells the same story from the other side. At low queue depths, every configuration answers 4K reads in 0.13 to 0.15ms, which is memory-and-wire territory, not disk. Response times scale predictably as the sweep deepens, and the NFS configurations stay well-behaved through moderate depths before climbing steeply once the box saturates at the extreme corners of the matrix. The practical read: ENAS has plenty of headroom for the concurrent small-file access an office generates.

4K Rand Write IOPS

Random 4K writes separate the configurations dramatically. Async NFS peaks at 9.6K IOPS (10.1K with cache), respectable for a RAID-Z2 pool of spinning disks. Force synchronous writes, though, and throughput collapses to a few hundred IOPS, with mean latencies in the tens of milliseconds; every commit is waiting on the pool. UniFi Drive does not expose a dedicated ZFS write-log (SLOG) device, which is the usual fix for exactly this behavior, and the ENAS SSD cache is read-only, so there is nothing in the platform today to absorb synchronous writes. iSCSI landed at 4.5K IOPS in the base configuration and 15.2K with the cache populated, an unexpectedly large spread for a read cache, so treat the delta between those two runs with some caution.

4K Rand Write Latency

The write-latency chart mirrors the IOPS results: async NFS holds sub-millisecond response at the low end of the sweep and degrades gracefully, while the sync configurations start near 20ms and climb from there. Nothing here is surprising for ZFS on hard drives without a log device. Still, shops planning sync-write-heavy workloads, databases, or NFS datastores with sync enforced should look at the async and iSCSI numbers and plan accordingly.

1M Seq Read GB/s

Large sequential reads are the ENAS headline result. Async NFS peaked at 4,170 MB/s, and all four NFS configurations held above 4,100 MB/s, comfortably beyond what a single 25GbE link can carry, confirming the system can drive both SFP28 ports at once. For a $3,999 box full of hard drives, north of 4GB/s of sustained read throughput is a legitimately strong showing that covers backup verification, media pulls, and multi-client streaming with room to spare. iSCSI sequential reads are the sore spot: the target peaked at just 257MB/s (227MB/s with cache) and degraded as load increased, behavior consistent with the target oversaturating rather than throttling cleanly. This looks like the kind of issue a firmware release resolves; until then, iSCSI on ENAS is better suited to random-access workloads than large sequential streams.

1M Seq Read Latency

Sequential read latency stays orderly on NFS, opening around 1.7ms and scaling with queue depth as expected. The iSCSI curve is the outlier, with response times an order of magnitude higher at equivalent load, which is the same oversaturation signature seen in the throughput chart.

1M Seq Write GB/s

Sequential writes over async NFS reached 2,567MB/s (2,654MB/s with cache), roughly the ceiling we would expect the RAID-Z2 pool itself to sustain. The sync NFS configurations tell the same story as the 4K results, topping out at 137 to 142MB/s. iSCSI writes fared far better than iSCSI reads at 1,432MB/s, further evidence that the read-path anomaly is a software issue rather than a hardware limit.

1M Seq Write Latency

Write latency closes out the sweep with no surprises: async NFS holds low single-digit milliseconds at sane depths, sync NFS lives in the tens of milliseconds, and iSCSI sits in between. Taken together, the data says ENAS is an excellent sequential machine and a capable async random performer, with sync-write workloads being the one area that demands planning.

Conclusion

The hardware value here is real. For $3,999, ENAS delivers 16 open bays, ZFS with ECC memory, dual 25GbE on board, redundant power, and expansion ports that will eventually take it past a petabyte, a combination the incumbents do not match at this price without add-on cards and, in some cases, drive-compatibility strings attached. Our testing backs the core proposition: over 4.1GB/s of sequential NFS read throughput, 2.6GB/s writes, and roughly 90K IOPS of cached 4K random reads is more performance than most small businesses will ever pull through it.

The decision comes down to software. Shops that live in Synology or QNAP’s application ecosystems, containers, surveillance, turnkey backup suites, will find UniFi Drive spartan by comparison, and virtualization builders should weigh the iSCSI sequential-read anomaly and the sync-write penalty before pointing a datastore at it. But that is not really who this box is for. For the SMB with a moderate or large UniFi estate that needs easy, reliable file and block storage, managed from the console it already knows, with snapshots, replication, cloud backup, and directory integration included and no license invoice ever arriving, ENAS is the most credible storage product Ubiquiti has shipped, and the first one the incumbents need to take seriously. Given the pace of UniFi Drive releases since our unit arrived, we expect the software gap to keep narrowing, and we will revisit the platform as immutable snapshots and the expansion shelves land.

Product Page – Ubiquiti ENAS

Ubiquiti Enterprise NAS (ENAS) (affiliate link)

The post Ubiquiti Enterprise NAS Review: 16 Bays, Dual 25GbE, and ZFS for $3,999 appeared first on StorageReview.com.