Random Read Performance

Our first test of random read performance uses very short bursts of operations issued one at a time with no queuing. The drives are given enough idle time between bursts to yield an overall duty cycle of 20%, so thermal throttling is impossible. Each burst consists of a total of 32MB of 4kB random reads, from a 16GB span of the disk. The total data read is 1GB.

Burst 4kB Random Read (Queue Depth 1)

The Optane SSD 900P doesn't break the record for QD1 random reads, but only because we've also tested the 32GB Optane Memory M.2, which is about two microseconds faster on average for each 4kB read. The Optane SSD 900P is still about 7 times faster than any flash-based SSD.

Our sustained random read performance is similar to the random read test from our 2015 test suite: queue depths from 1 to 32 are tested, and the average performance and power efficiency across QD1, QD2 and QD4 are reported as the primary scores. Each queue depth is tested for one minute or 32GB of data transferred, whichever is shorter. After each queue depth is tested, the drive is given up to one minute to cool off so that the higher queue depths are unlikely to be affected by accumulated heat build-up. The individual read operations are again 4kB, and cover a 64GB span of the drive.

Sustained 4kB Random Read

When longer transfers and higher queue depths come into play, the Optane SSD 900P passes the Optane Memory M.2 and remains more than 6 times faster for random reads than any flash-based SSD.

Both Optane devices more or less level off at queue depths of 8 or higher. The Optane SSD 900P saturates at about 1800 MB/s while the Optane Memory tops out around 1300 MB/s. The Samsung 960 PRO 2TB hasn't caught up by QD32, and doesn't surpass the QD1 random read performance of the Optane SSD until the Samsung reaches a queue depth of about 8.

Random Write Performance

Our test of random write burst performance is structured similarly to the random read burst test, but each burst is only 4MB and the total test length is 128MB. The 4kB random write operations are distributed over a 16GB span of the drive, and the operations are issued one at a time with no queuing.

Burst 4kB Random Write (Queue Depth 1)

The burst random write performance of the Optane SSD 900P is slightly higher than the Intel SSD 750 1.2TB, and about 14% faster than Samsung's fastest.

As with the sustained random read test, our sustained 4kB random write test runs for up to one minute or 32GB per queue depth, covering a 64GB span of the drive and giving the drive up to 1 minute of idle time between queue depths to allow for write caches to be flushed and for the drive to cool down.

Sustained 4kB Random Write

With higher queue depths in play, the Optane SSD 900P scales up faster than the Intel SSD 750 1.2TB, leaving the Optane SSD with a 7-10% lead over the Samsung 960s and Intel 750.

Samsung's 960 PROs and the larger 960 EVO all trail slightly behind the Optane SSD's random write performance for queue depths 1 to 4, then the Samsung drives level off and leave the Optane SSD with a substantial performance advantage at high queue depths. The Intel 750 is slightly faster at QD1 and QD2, but saturates at an even lower performance level than the Samsung 960s.

AnandTech Storage Bench - Light Sequential Performance
Comments Locked

205 Comments

View All Comments

  • melgross - Friday, October 27, 2017 - link

    You’re not as objective as you proclaim. If you were, you’d be more in line with what the conslucion says, which is that for many things there isn’t much of an advantage, but for others there is, and most of the reas\OSs aren’t the fault of this, but rather, old concepts in storage.
  • ddriver - Friday, October 27, 2017 - link

    So in order to be objective, I will have to echo what AT - a heavily pro-intel biassed website says about it? Yep, that sounds legit :)

    It would seem you confuse objectivity for conformity.
  • r0gue6 - Friday, October 27, 2017 - link

    I can tell you are extremely bias simply because you use ignorant words like "hypetane" unironically.
  • ddriver - Friday, October 27, 2017 - link

    Hypetane is perfectly suited. The product turned out to be 90% hype, and it also rhymes. What more could anyone possibly want?

    And calling things for what they are is the very essence of objectivity.
  • shabby - Friday, October 27, 2017 - link

    It is 1000 times faster... at die level, pcie and drivers show it down. /intel pr pro
  • ddriver - Friday, October 27, 2017 - link

    Sure, if you say so. I mean it is obviously saturating and exceeding the PCIE bandwidth and crippled by the PCIE latency.

    Oh wait, it isn't. Maybe it is intel's controller then. Who knows. I mean aside from you ;)
  • lmcd - Friday, October 27, 2017 - link

    Ironic, because you aren't the king of anything.
  • voicequal - Saturday, October 28, 2017 - link

    It's pretty clear Intel's performance claims were speaking of the cell level (NAND vs 3D Xpoint), not the system level (SSD, SYSmark, etc).
  • Drumsticks - Friday, October 27, 2017 - link

    Not that anybody will ever convince you that anything Intel can do is good, but here is a graph of how much endurance the 480GB Optane SSD has in terms of TBW compared to the other Pro SSDs.

    https://www.pcper.com/files/imagecache/article_max...
  • ddriver - Friday, October 27, 2017 - link

    Cool, even if not real-world maxing out but intel's claim, I am willing to assume that intel won't lie on the spec sheet of an actual product.

    But then again, I am going to refer you to this little gem:

    http://images.hardwarecanucks.com/image/akg/Storag...

    The rated number is still almost 50 times less than what intel claimed officially. 45 times less is pretty significant IMO. Imagine getting a job where they promise you 20k $ a month, and end up paying you 450$ instead. Not cool. In light of that, I wouldn't say my criticism is ungrounded.

    For me personally, 20 times better endurance than NAND is pretty good, good enough to justify the purchase even at the present price.

    Now if only intel started out with a 20x claim instead of the 1000x claim, I wouldn't have any legit reason to bash that product. And maybe if more call intel on their BS, they just might cut it. And I don't think that will be a bad thing.

Log in

Don't have an account? Sign up now