Performance Comparisons – Client QD Weighted

These results attempt to simplify things by focusing on what really matters – the Queue Depths that folks actually see when using these products. A dimension is eliminated from the previous charts by applying a weighted average to those results. The weights were derived from trace recordings of moderate to heavy workloads, which still ended up running closer to QD=1-2 even on a slower SATA SSD. The intent here is to distill the results into something for those wanting 'just the facts' to grab and go when making their purchasing decisions. Don't be alarmed by the low figures. Remember, these are low queue depths – the place where these SSDs actually operate when in use by those not just running benchmarks all day!

qd-rnd.png

We already know the random write performance is similar, so let us focus on the blue bars here. Optane's random read performance is simply outstanding. We are talking nearly 5x the IOPS at these lower queue depths (where IOPS actually matters).

qd-seq.png

The Samsung 960 PRO is able to best the SSD 900P's in straight line sequential reads, but the same can not be said for the 960 EVO, which comes in at a tie due to its poorer low QD performance seen earlier in this review. Remember though, reads and writes are rarely isolated. There's usually something going on in the background, and *those* results fall on the next page.

c-rs.png

We don't normally include this chart because many of the lines overlap each other, making it hard to read. Not so this time. It's pretty easy to separate Optane from the rest here. What we are looking at is random performance across a 'read sweep', focusing on queue depths seen in client use, where the read percentage of a mixed workload increases from left to right. Note how with only 10-20% reads mixed in, the NAND SSD performance has already fallen significantly, while the SSD 900P's are climbing!

« PreviousNext »