Direct:
Breakdown for those w/o access.
The solution has three dimensions:
- Workload-aware decision engine – analyzes the characteristics of a workload to pick the best config
- Architecture synthesis – integrates five different RAID architectures into one system
- Performance-Reliability-Cost cube model – a model to evalute and balance the factors
Result:
- P95 latency reduced by 22%
- 2x mean-time-to-data-loss (MTTDL)
- TCO lowered by 18%
Reddit: https://ift.tt/0aRf4HT