Sequential Performance Vs. Transfer Size
We just looked at 128 KB sequential performance in Iometer, but that only shows the performance of each drive at a queue depth of one. In the real world, there are situations where you'll see higher queue depths, even with a performance-oriented SSD. That's why it's important to also factor in those higher queue depths into our performance analysis.
This time, we're using ATTO to test the sequential reads/writes over 2 GB using a queue depth of two. Why only a queue depth of two? Even when you're pushing the envelope, operations complete so much faster on an SSD that queue depths higher than two or three are far less common on an average desktop.
The other reason to use ATTO is its ability to easily test different transfer sizes. While 128 KB is the standard block size for measuring sequential performance, there are situations where you're going to deal with smaller or larger transfer sizes. Block sizes are often larger than 1 MB. On the lower end of the spectrum, DLLs and file dependencies are often 4 KB or less.
As we look at a slightly higher queue depth, larger-capacity SSDs no longer offer superior 128 KB performance to lower capacity SSDs in sequential reads. The 512 GB m4 achieves about 210 MB/s, but the 64 GB m4 hits somewhere close to 260 MB/s.
Sequential write performance is a more clear-cut case where more capacity translates into additional performance. Here we see the 512 and 256 GB m4s at the top of their game, achieving as much as 275 MB/s. The RealSSD C300 falls slightly behind at 240 MB/s, while the 128 GB m4 drops to about 190 MB/s.
Armed with half of the capacity, the 64 GB model performs noticeably worse, its sequential writes topping out at about 100 MB/s. However, that's a lot better than what we saw from Iometer with a queue depth of one. Now, we're seeing better numbers from the SSD than Seagate's 500 GB Momentus 5400.6.