4 KB Random Performance: Throughput
Our Storage Bench v1.0 mixes random and sequential operations. However, it's still important to isolate 4 KB random performance, because that's such a large portion of what you're doing on a day-to-day basis. Right after Storage Bench v1.0, we subjected the drives to Iometer to test random 4 KB performance. But why specifically 4 KB?
When you open Firefox, browse multiple Web pages, and write a few documents, you're mostly performing small random read and write operations. The chart above comes from analyzing Storage Bench v.10, but it epitomizes what you'll see when you analyze any trace from a desktop computer. Notice that close to 70% of all of our accesses are eight sectors in size (512 bytes per sector, thus 4 KB).
We're restricting Iometer to test an LBA space of 16 GB because a fresh install of a 64-bit version of Windows 7 takes up nearly that amount of space. In a way, this examines the performance that you would see from accessing various scattered file dependencies, caches, and temporary files.
If you're a typical PC user, it's important to examine performance at a queue depth of one. However, the RevoDrive 3 X2 is anything but a typical SSD. OCZ envisions high queue depth use. That really shows up in the random read performance line graph.
At lower queue depths, the 256 GB and 512 GB m4s, 256 GB C300, and even the older RevoDrive X2 outpace the RevoDrive 3 X2. Once you move up to queue depths above eight, OCZ's latest PCIe SSD leaves everything else in the dust as it approaches 700 MB/s (~175 000 IOPS).
If you're a hard drive owner, this benchmark is really a worst-case scenario. Seagate's 5400.6 only achieves random reads of 0.6 MB/s random read rate, 150x slower than the 64 GB m4.
The RevoDrive 3 X2 performs better in random writes. Even at a queue depth of one, its write rate of 224.6 MB/s is only outpaced by the 256 GB and 512 GB m4s. However, the margin of difference is very small among the top five SSDs. The RevoDrive 3 X2 really starts to make its stand once you look at queue depths above four. At a queue depth of eight, you're almost pushing 800 MB/s (~200 000 IOPs).
