Demartek Publishes Article on Latency and IOPS on All-flash Arrays
For storage systems, a test lab will typically concern itself with three basic metrics of performance:
- Throughput measured in megabytes per second (MBps)
- Latency measured in milliseconds or microseconds in addition to other metrics from the application host server.
In testing all-flash arrays, some interesting results become apparent right away. Compared to hard disk drive (HDD)-based arrays, there is a significant difference in the three basic metrics and performance consistency. Although it is workload dependent, all-flash arrays generally have more consistent overall performance than HDD-based arrays. This is especially apparent in the latency measurements.
Many workloads—such as online transaction processing database, VDI and Web server workloads—benefit from latency improvements. When these workloads are moved to all-flash arrays, the end-user experience is dramatically improved because response time is reduced, and the increased performance is consistent.
Many all-flash arrays can reduce average latency to less than 1 millisecond, depending on the workload.