Mixed Random Performance

Our test of mixed random reads and writes covers mixes varying from pure reads to pure writes at 10% increments. Each mix is tested for up to 1 minute or 32GB of data transferred. The test is conducted with a queue depth of 4, and is limited to a 64GB span of the drive. In between each mix, the drive is given idle time of up to one minute so that the overall duty cycle is 50%.

Mixed 4kB Random Read/Write

The mixed random I/O performance of the Intel Optane SSD 900P is completely unmatched. The small Optane Memory M.2 is the second fastest drive we've tested, and the fastest flash-based SSD is only a third as fast as the Optane SSD overall.

At the very end of the test, when the workload shifts to pure random writes, Samsung's fastest SSDs are able to batch the writes and dramatically improve throughput, almost enough to catch up to the slowest speed the Optane SSD hits during this test.

Mixed Sequential Performance

Our test of mixed sequential reads and writes differs from the mixed random I/O test by performing 128kB sequential accesses rather than 4kB accesses at random locations, and the sequential test is conducted at queue depth 1. The range of mixes tested is the same, and the timing and limits on data transfers are also the same as above.

Mixed 128kB Sequential Read/Write

The Intel Optane SSD 900P doesn't quite dominate the mixed sequential I/O test to the extent that it trounced the competition in the mixed random I/O test, but it still breaks the record with a 40% higher average throughput than the fastest flash-based SSD.

The Optane SSD 900P's performance wavers a bit as the workload changes, but the general trend is a gradual reduction in performance as the proportion of writes increases. The flash-based SSDs tend to show an sharper decline in performance during the first half of the test, and the good ones recover most of that performance over the course of the second half. But the low performance in the middle of the test brings the averages way down and leaves the Optane SSD alone at the top.

Sequential Performance BAPCo SYSmark 2014 SE
Comments Locked

205 Comments

View All Comments

  • Billy Tallis - Friday, October 27, 2017 - link

    The main power meter failed before the Optane drive arrived. The whole-system power meter was unaffected.
  • lmcd - Friday, October 27, 2017 - link

    I can't trust your comments either anymore, @jjj. But who's counting?
  • wookie monster - Friday, October 27, 2017 - link

    Why no write consistency test? I was able to experiment with a prototype Optane drive, and I found that running a long-haul randomly-ordered write test on the Optane drive was substantially faster than the fastest available flash-based SSDs.
  • Billy Tallis - Friday, October 27, 2017 - link

    I plan to do a lot more to this drive, including steady-state random write comparisons against consumer and enterprise SSDs. There just wasn't time to include more tests in this article. I've only had the drive for six days.
  • wookie monster - Friday, October 27, 2017 - link

    I look forward to the results, thanks!
  • willis936 - Friday, October 27, 2017 - link

    I also really like the performance over time plots. An investigation into power and thermals (and potential throttling) would be interesting.
  • takeshi7 - Friday, October 27, 2017 - link

    It's pretty obvious that once an SSD is installed in the system, game load times are limited by the CPU, not the storage. This is why this Optane drive won't load games significantly faster than a SATA SSD (especially when you consider the price increase). Can Anandtech please test how different CPUs affect game loading times?
  • Scannall - Friday, October 27, 2017 - link

    For the vast majority out there, this seems like a solution in search of a problem. As expensive as it is, you'd be better off raiding a couple nvme drives and calling it good.
  • ddriver - Friday, October 27, 2017 - link

    Nope, raid-ing will in no way improve their weak spots - random and low queue depth access. It will only boost sequential and high QD performance, which is already superior to that of hypetane.
  • citrix13 - Friday, October 27, 2017 - link

    Thank you ddriver for your objective, unbiased and logical observations.
    Many in here cannot comprehend what you have been saying which is unfortunate.
    Your core point is that Intel promised 1000x performance with Intel Optane
    Intel did not deliver 1000x performance, they gave orders of magnitude less
    Also, i take note that you praised the drives endurance and low queue depth performance and said you may buy it.
    Commendation on being able to call a spade a spade, but still being able to see it strengths

Log in

Don't have an account? Sign up now