Version and / or Patch Used: 3.05
ATTO is a timeless benchmark used to provide manufacturers with data used for marketing storage products. When evaluating ATTO performance we focus on the drive's performance curve.
Full performance is achieved at 32KB transfers. We can see that unpatched delivers significantly better small-file performance than patched.
Overall, Samsung's 970 EVO displays the best performance curve of the bunch. The 905P's unpatched performance curve is better than the 900P and the 905P tops out at a higher rate. Patched vs. unpatched performance is equal on the low-end and on the top end. The important middle-ground is where patched performance takes a big hit.
Overall, the 905P slightly outperforms the 900P, but both are edged out for the top spot by the 960 Pro. There are other SSDs in our test pool that achieve higher transfer rates on the top end, but their performance curve is inferior.
Anvil Storage Utilities
Version and / or Patch Used: 1.1.0
Anvil's Storage Utilities is a storage benchmark designed to measure the storage performance of SSDs. The Standard Storage Benchmark performs a series of tests; you can run a full test or just the read or write test, or you can run a single test, i.e. 4K QD16. When evaluating performance with Anvils, we focus on total score. When evaluating NVMe SSDs we are typically looking for a minimum total score of over 10K. We place a greater importance on read performance than write performance.
Unpatched performance results in a net score of over 2K better than when patched.
The 905P delivers better overall performance than the 900P as evidenced by these results. Patched performance is significantly lower than unpatched, but its still far better than any of the non-Optane SSDs that comprise our test pool. Keep in mind that only the 905P is running on a patched system.
(Anvil) Read IOPS through Queue Depth Scale
Unpatched; we are very close to matching factory max random read IOPS. With an OS and 75% full we can't quite get there at QD32.
As this chart clearly demonstrates, Optane SSDs are in a class of their own when reading random data. We are seeing 5x the performance of our flash-based contenders at QD1. This is where performance matters most, and this is where Optane delivers like no other. Here again, we see the 905P outperforming the 900P.
(Anvil) Write IOPS through Queue Scale
With a partition on the drive, random write IOPS at QD32 fall short of factory specs. We are seeing a more than 75K IOPS advantage for unpatched vs. patched. That may not sound like a lot with so many IOPS in play, but 75K is about a SATA SSD's worth of performance difference which is substantial.
This is a tough one to call. The SanDisk Extreme Pro delivers the best low queue depth performance, the 970 EVO the best at queue depths above 4. Because we value low queue performance more than anything else, we give the win to the Extreme Pro. Keep in mind that while random write performance is important, it is not anywhere near as important as random read.
Last updated: Nov 15, 2019 at 01:16 pm CST
PRICING: You can find products similar to this one for sale below.
United States: Find other tech and computer products like this over at Amazon.com
United Kingdom: Find other tech and computer products like this over at Amazon.co.uk
Australia: Find other tech and computer products like this over at Amazon.com.au
Canada: Find other tech and computer products like this over at Amazon.ca
Deutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de
- Page 1 [Introduction, Drive Specifications, Pricing & Availability]
- Page 2 [Drive Details]
- Page 3 [Test System Setup & Drive Properties]
- Page 4 [Synthetic Benchmarks - ATTO & Anvil Storage Utilities]
- Page 5 [Synthetic Benchmarks - CrystalDiskMark & AS SSD]
- Page 6 [Benchmarks (OS) - Vantage, PCMark 7, PCMark 8 & More]
- Page 7 [Benchmarks (Secondary) - IOPS, Response & Transfer Rate]
- Page 8 [Benchmarks (Secondary Volume) - PCMark 8 Extended]
- Page 9 [Benchmarks 70/30 Mixed Workload & Sustained Seq. Write]
- Page 10 [Maxed-Out Performance (MOP)]
- Page 11 [Final Thoughts]