Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal fragmentation. The reason we do not have consistent IO latency with SSDs 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 (Logical Block Addresses) 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 dropdown selections 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.

Corsair Neutron XT 240GB
Default
25% Over-Provisioning

Performance consistency has never been Phison's biggest strength and that continues to be the case with the S10 controller. The consistency is actually worse compared to the older S8 controller (i.e. Corsair Force LS) because the variance in performance is so high. I'm pretty sure the issue lies in Phison's garbage collection architecture as it doesn't seem to give enough priority for internal garbage collection, which results in a scenario where the drive has to stop for very short periods of time (milliseconds) to clean up some blocks for the IOs in the queue. That is why the performance frequently drops to ~1,500 IOPS, but on the other hand the drive may be pushing 70K IOPS the second after. Even adding more over-provisioning doesn't produce a steady line, although the share of high IOPS bursts is now higher. 

For average client workloads, this shouldn't be an issue because drives never operate in steady-state and IOs tend to come in bursts, but for users that tax the storage system more there are far better options on the market. I'm a bit surprised that despite having more processing power than its predecessors, the S10 can't provide better IO consistency. With three of the four cores dedicated to flash management, there should be plenty of horsepower to manage the NAND even in steady-state scenario, although ultimately no amount of hardware can fix inefficient software/firmware. 

Corsair Neutron XT 240GB
Default
25% Over-Provisioning

 

Corsair Neutron XT 240GB
Default
25% Over-Provisioning


TRIM Validation

To test TRIM, I filled the drive with sequential 128KB data and proceeded with a 30-minute random 4KB write (QD32) workload to put the drive into steady-state. After that I TRIM'ed the drive by issuing a quick format in Windows and ran HD Tach to produce the graph below.

And TRIM works as expected.

Introduction, The Drives & The Test AnandTech Storage Bench 2013
POST A COMMENT

56 Comments

View All Comments

  • hojnikb - Monday, November 17, 2014 - link

    At this point, we have a better chance of finding Loch Ness Monster than getting sf3700 :) Reply
  • eddieobscurant - Monday, November 17, 2014 - link

    epic !!! Reply
  • glugglug - Monday, November 17, 2014 - link

    Any endurance testing done?

    I have a Neutron 512GB (not GTX) and am pretty disppointed at what the wear leveling indicator from SMART data is telling me..... it looks like the 512GB Neutron has an expected write endurance of only 100TB?
    Reply
  • Kristian Vättö - Monday, November 17, 2014 - link

    I haven't done any endurance testing, but Corsair is rating the Neutron XT at 124TB (according to The Tech Report). Note that the SMART data isn't usually an indication of an expected failure as the threshold has been set by the manufacturer for warranty reasons, so they can determine whether you've exceeded the endurance rating. In that light 100TB is pretty good because most drives are rated at around 70TB or so. Reply
  • Joepublic2 - Monday, November 17, 2014 - link

    I'm disappointed to see them dropping the 5 year warranty on the Neutron line. I bought a Neutron and a Neutron GTX for myself and a few dozen for various clients primarily due to the 5 year warranty. Expected endurance figures mean nothing to me as a user if the company isn't willing to back it up with a appropriate warranty. Reply
  • Kristian Vättö - Tuesday, November 18, 2014 - link

    Actually, there was an error in the reviewer's guide, so the warranty is indeed 5 years similar to the original Neutron series. I've updated the article as well. Reply
  • glugglug - Sunday, November 23, 2014 - link

    That is very reassuring.

    One other disturbing thing is the raw read error rate reported in SMART. It starts out looking good after a cold boot, then gradually drops to a value of "1" as the system gets a higher uptime. I think this is actually calculated wrong, being a direct mapping from the raw value, which is always identical to the "Soft ECC Correction Rate" raw value (sitting at ~250,000 after about a month of uptime), instead of being calculated based on the rate of change in that raw value like its supposed to be. Corsair SSD Toolbox labels that counter as "informational only", while CrystalDiskInfo, Stablebit Scanner, & other tools complain when the nominal value crosses the apparent manufacturer suggested threshold of "6" after a week or so.

    Restarting Windows does NOT reset the counter -- it gets reset only by a cold boot.
    Reply
  • extide - Monday, November 17, 2014 - link

    That wear leveling indicator has very little to do with ACTUAL flash lifetime. It is mostly to do with giving the manufacturer the ability to determine when warranty has expired. Reply
  • sonicmerlin - Monday, November 17, 2014 - link

    It's really kinda cool these things have quad core processors now.

    On a side note, does anyone know when someone other than Samsung is going to release 3D NAND?
    Reply
  • Kristian Vättö - Tuesday, November 18, 2014 - link

    IMFT (i.e. Micron and Intel) is next year and so is SK Hynix. Toshiba/SanDisk is H1'16. Reply

Log in

Don't have an account? Sign up now