Performance Consistency

Performance consistency tells us a lot about the architecture of these SSDs and how they handle internal defragmentation. 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 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.

  Transcend SSD340 JMicron JMF667H (Toshiba NAND) JMicron JMF667H (IMFT NAND) Samsung SSD 840 EVO mSATA Crucial MX100
Default
25% OP

Right from the start, things do not look too promising. Compared to the reference design with the same IMFT NAND, the IO consistency is considerably lower. The reference design manages around 1,000 IOPS minimum, whereas in the SSD340 the minimum performance is around 300 IOPS. Increasing the over-provisioning helps by a bit but the consistency is still poor compared to the other value drives (like the MX100). The older firmware definitely isn't doing Transcend any favors here – quite the opposite in fact.

  Transcend SSD340 JMicron JMF667H (Toshiba NAND) JMicron JMF667H (IMFT NAND) Samsung SSD 840 EVO mSATA Crucial MX100
Default
25% OP

 

  Transcend SSD340 JMicron JMF667H (Toshiba NAND) JMicron JMF667H (IMFT NAND) Samsung SSD 840 EVO mSATA Crucial MX100
Default
25% OP

TRIM Validation

To test TRIM, I first filled all user-accessible LBAs with sequential data and continued with torturing the drive with 4KB random writes (100% LBA, QD=32) for 30 minutes. After the torture I TRIM'ed the drive (quick format in Windows 7/8) and ran HD Tach to make sure TRIM is functional.

And it is. 

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

56 Comments

View All Comments

  • hojnikb - Monday, August 04, 2014 - link

    Could it be possible to flash newer refrence firmware to it ?

    I know this have been done in the past, because i've flashed my Crucial V4 with newer firmware that crucial offered. I wonder if its the same with Jmicron
    Reply
  • solnyshok - Monday, August 04, 2014 - link

    Whenever I read about slow ssd, I wonder if "the one" that I need is going to happen. If manufacturers read this - here is the niche for HTPC. I need a silent movie/torrent storage. 1TB at $200-250 would be a good start. Performance/power saving aren't that important, as long as it can sustatin 130MB/s sequential read/write (bottleneck of home gigabit network) and 10,000+ iops (to avoid jerkiness of video playback during concurrent torrent seeding) Reply
  • ddriver - Monday, August 04, 2014 - link

    Most HDDs do 130 MB already. You can avoid "jerkiness" by tweaking buffering/caching for playback and torrents respectively. Reply
  • leexgx - Tuesday, August 05, 2014 - link

    can you fix the Crucial MX100 OP results and lower the IOPS to to 30k

    as the Crucial MX100 is doing some very odd results on the 25% OP 32QD 4K Write (the last one seems an error doing 60k IOPS as its way past 50K on that last test) 412GB is making the charts go to 50k when most do not pass 25k
    Reply
  • leexgx - Tuesday, August 05, 2014 - link

    4K random Write 32Q 25% OP Crucial MX100 Reply
  • solnyshok - Tuesday, August 05, 2014 - link

    Thanks for all replies, I already have SSD (256GB 840EVO+2TB 2.5"HDD) combo in my htpc. I just wanted to go totally silent. My point was that somebody needs to pack 1TB of the cheapest flash on the cheapest controller for the lowest $/GB ratio. I appreciate replies about price-o-nomics of ssd making. If it is not feasible, oh, well, I can wait couple of years. Reply
  • MrSpadge - Monday, August 04, 2014 - link

    You hardly gain any cost benefit by making SSDs slow. That's why you don't see the value drives you're looking for. Sounds like a 5x00 rpm HDD together with some small to medium sized SSD for the torrents is your best solution. Assuming you don't offer your entire movie collection for torrenting, of course. Reply
  • hojnikb - Monday, August 04, 2014 - link

    If you tune your torrent client right, ANY 5400rpm hdd will be just fine. I know, because i'm using an old 2.5" drive in my nas and can do multiple torrents AND streaming with ease. Its just the right tuning. Reply
  • wolrah - Monday, August 04, 2014 - link

    It's not even tuning, just keeping preallocation on and keeping fragmentation down is enough. Reply
  • solnyshok - Tuesday, August 05, 2014 - link

    Thanks, if that's the case, I just need to wait a bit (3 years?), until 1TB becomes cheap Reply

Log in

Don't have an account? Sign up now