Single page Print

Sustained and scaling I/O rates
Our sustained IOMeter test hammers drives with 4KB random writes for 30 minutes straight. It uses a queue depth of 32, a setting that should result in higher speeds that saturate each drive's overprovisioned area more quickly. This lengthy—and heavy—workload isn't indicative of typical PC use, but it provides a sense of how the drives react when they're pushed to the brink.

We're reporting IOps rather than response times for these tests. Click the buttons below the graph to switch between SSDs.

To show the data in a slightly different light, we've graphed the peak random-write rate and the average, steady-state speed over the last minute of the test.

The Reactor 1TB's peak random write rate is pretty speedy. The 850 EVO 1TB and Vector 180 960GB stand head and shoulders above it, but that stiff competition comes from drives in a completely different price class. The Reactor doesn't have any fancy pseudo-SLC tricks like the EVO's TurboWrite, either. On the steady-state side of our testing, the Reactor inexplicably trails the BX100 500GB by a large margin, despite sharing the same controller and NAND. Perhaps Micron knows how to coax more mojo out of its MLC than it cares to share with third-party OEMs.

Our final IOMeter test examines performance scaling across a broad range of queue depths. We ramp all the way up to a queue depth of 128. Don't expect AHCI-based drives to scale past 32, though—that's the maximum depth of their native command queues.

For this test, we use a database access pattern comprising 66% reads and 33% writes, all of which are random. The test runs after 30 minutes of continuous random writes that put the drives in a simulated used state. Click the buttons below the graph to switch between the different drives. And note that the P3700 plot uses a much larger scale.

Unlike many of the SATA drives we've looked at recently, the Reactor 1TB's performance does scale as queue depth increases. There's an almost 100% increase between QD1 and the higher queue depths. That's nowhere near the astronomical scaling of an Intel PCIe drive, but it's certainly better than the completely flat scaling performance we often see. Below, we plot it against the scaling results of other drives to provide context.

The Reactor's scaling curve isn't markedly different from the 850 EVO's. As usual, we threw one of OCZ's Barefoot-controlled drives into the mix to put the others to shame. Next up, we'll look at real-world performance with some basic file I/O.