Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal defragmentation. The reason we don’t have consistent IO latency with SSD is because inevitably all controllers have to do some amount of defragmentation or garbage collection in order to continue operating at high speeds. When and how an SSD decides to run its defrag or cleanup routines directly impacts the user experience as inconsistent performance results in application slowdowns.

To test IO consistency, we fill a secure erased SSD with sequential data to ensure that all user accessible LBAs have data associated with them. Next we kick off a 4KB random write workload across all LBAs at a queue depth of 32 using incompressible data. The test is run for just over half an hour and we record instantaneous IOPS every second.

We are also testing drives with added over-provisioning by limiting the LBA range. This gives us a look into the drive’s behavior with varying levels of empty space, which is frankly a more realistic approach for client workloads.

Each of the three graphs has its own purpose. The first one is of the whole duration of the test in log scale. The second and third one zoom into the beginning of steady-state operation (t=1400s) but on different scales: the second one uses log scale for easy comparison whereas the third one uses linear scale for better visualization of differences between drives. Click the buttons below each graph to switch the source data.

For more detailed description of the test and why performance consistency matters, read our original Intel SSD DC S3700 article.

  SanDisk Extreme Pro SanDisk Extreme II Intel SSD 730 Intel SSD 530 OCZ Vector 150
Default
25% Spare Area

Similar to the Extreme II, the IO consistency is just awesome. SanDisk's firmware design is unique in the sense that instead of pushing high IOPS at the beginning, the performance drops close to 10K IOPS at first and then rises to over 50K and stays there for a period of time. The higher the capacity, the longer the high IOPS period: the 960GB Extreme Pro takes ~800 seconds before the IOPS drops to 10K (i.e. the drive reaches steady-state). I do not know why SanDisk's behavior is so different (maybe it has something to do with nCache?) but it definitely works well. Furthermore, SanDisk seems to be the only manufacturer that has really nailed IO consistency with a Marvell controller because Crucial/Micron and Plextor have had some difficulties and their performance is not even close to SanDisk.

However, I would not say that the Extreme Pro is unique. Both Intel SSD 730 and OCZ Vector 150 provide the same or even better performance at steady-state, and with added over-provisioning the difference is even more significant. That is not to say that the Extreme Pro is inconsistent, not at all, but for a pure 4KB random write workload there are drives that offer (slightly) better performance.

  SanDisk Extreme Pro SanDisk Extreme II Intel SSD 730 Intel SSD 530 OCZ Vector 150
Default
25% Spare Area

 

  SanDisk Extreme Pro SanDisk Extreme II Intel SSD 730 Intel SSD 530 OCZ Vector 150
Default
25% Spare Area

 

TRIM Validation

To test TRIM, I filled the drive with sequential data and proceeded with 60 minutes of 4KB random writes at queue depth of 32. I measured performance with HD Tach after issuing a single TRIM pass to the drive.

TRIM works for sure as the write speed is at steady 400MB/s.

.

Introduction, The Drives & The Test AnandTech Storage Bench 2013
Comments Locked

85 Comments

View All Comments

  • Kristian Vättö - Tuesday, June 17, 2014 - link

    Unfortunately the tool we use doesn't allow the LBA range to be limited.
  • uruturu - Wednesday, June 18, 2014 - link

    :( I think would have been a head-to-head between Sandisk Extreme Pro, Samsung 840 Pro and Corsair Neutron GTX...
  • sheh - Tuesday, June 17, 2014 - link

    It's good to see also the lower capacity model performing well. I hope other manufacturers would do the same. But too bad there's no 120GB.

    "SanDisk is betting that its target users will ... or the endurance limit will be hit after the warranty runs out"

    Is the warranty not 10 years or 80TB, whichever comes first?
  • Kristian Vättö - Tuesday, June 17, 2014 - link

    Yes. The warranty will be voided if you exceed the 80TB write limit.
  • hojnikb - Tuesday, June 17, 2014 - link

    wow, i never though there are products with 32!!! stacked dies.
    is there any specific product, that uses so much dies ?
  • prophet001 - Tuesday, June 17, 2014 - link

    I was wondering why there was no mention of block size. Do you set the block size on these drives? If so, does it make a difference in the performance?
  • Demon-Xanth - Tuesday, June 17, 2014 - link

    What's going to keep PCIe SSD adoption low is the sheer number of systems that you can't just drop it in and boot in it's current state. What it will likely take is for the drives to ship with a legacy mode that says to BIOS "yeah... I'm a SATA RAID controller with a bunch of drives attached...". The latest EFI updates for NVE and Windows 8.1 are a great step forward though.
  • zuiop - Tuesday, June 17, 2014 - link

    Can somebody explain to me why in most SSD reviews the lack of encryption support is criticised? Of course you want encryption for your data, but why in the drive? It's a black box implementation that you can't fully trust or update, unlike the plenty of open source software solutions. Especially when talking about high performance drives, I'd expect the CPU to have hardware acceleration for encryption, so speed shouldn't be an issue afaict. Can anybody shed some light on this?
  • Kristian Vättö - Tuesday, June 17, 2014 - link

    Software encryption adds unnecessary overhead and thus degrades performance:

    http://www.anandtech.com/show/6891/hardware-accele...

    I think I need to do an article that better describes the usefulness of TCG Opal because there are some security advantages as well. I admit that we probably haven't done the best to fully address this, but hopefully we can still fix it :)
  • zuiop - Tuesday, June 17, 2014 - link

    Thanks for your reply. I'd love to see the article. I couldn't see from that linked article whether the CPU has AES NI or equivalent, my point is that I can't see why encrypting the data with such a CPU is inferior to drive side encryption. I'm looking forward to the article!

Log in

Don't have an account? Sign up now