AnandTech Storage Bench - The Destroyer

The Destroyer is an extremely long test replicating the access patterns of very IO-intensive desktop usage. A detailed breakdown can be found in this article. Like real-world usage, the drives do get the occasional break that allows for some background garbage collection and flushing caches, but those idle times are limited to 25ms so that it doesn't take all week to run the test. These AnandTech Storage Bench (ATSB) tests do not involve running the actual applications that generated the workloads, so the scores are relatively insensitive to changes in CPU performance and RAM from our new testbed, but the jump to a newer version of Windows and the newer storage drivers can have an impact.

We quantify performance on this test by reporting the drive's average data throughput, the average latency of the I/O operations, and the total energy used by the drive over the course of the test.

ATSB - The Destroyer (Data Rate)

The 1TB models of the ADATA SX8200 Pro and HP EX950 go a long way toward catching up with other current high-end drives, but ultimately their overall performance on The Destroyer is inadequate. The 2TB EX950 doesn't even outperform last year's 1TB EX920.

ATSB - The Destroyer (Average Latency)ATSB - The Destroyer (99th Percentile Latency)

The average latency of the 1TB SM2262EN drives on The Destroyer is a significant improvement over their predecessors, but as with the average data rate it doesn't bring them up to the level of other recent high-end drives. The 99th percentile latency has actually regressed slightly from what the 1TB HP EX920 provided. The 2TB EX950 is slower in both metrics than the 1TB drives, but the 99th percentile latency score has improved from the pre-production firmware we tested last year.

ATSB - The Destroyer (Average Read Latency)ATSB - The Destroyer (Average Write Latency)

The 2TB HP EX950 has significantly worse average read latency than the 1TB models of the EX950 or EX920, but the average write latency is comparable to the 1TB EX950 and a clear improvement over the EX920. Regardless of capacity, the SM2262EN drives still have a lot of room for improving latency on The Destroyer.

ATSB - The Destroyer (99th Percentile Read Latency)ATSB - The Destroyer (99th Percentile Write Latency)

Overall, the 99th percentile read latency scores from the SM2262EN drives are closer to being competitive with other recent high-end drives than the 99th percentile write latencies, but both need to improve—especially the read QoS that has regressed slightly from the HP EX920. The SM2262 and SM2262EN drives have worse 99th percentile write latency on The Destroyer than the Crucial MX500, a SATA drive that is itself powered by a Silicon Motion controller.

For average latency, the 2TB EX950's weakness was on read operations, but when looking at 99th percentiles, it's writes that are a problem for the 2TB drive, though not as much as with the pre-production firmware.

ATSB - The Destroyer (Power)

The ADATA SX8200 Pro used significantly less power during The Destroyer than the HP EX950, putting the SX8200 Pro well ahead of the entire collection of SMI-based NVMe drives and close to the Phison E12-based Corsair MP510. WD and Toshiba remain the only ones to actually tie or surpass the power efficiency of SATA drives on this test. The 2TB EX950 unsurprisingly consumes a bit more energy than the 1TB model, thanks to having more memory to keep powered and due to firmware optimizations that are more ill-suited to this test than how the 1TB models behave.

SLC Cache Sizes & SYSmark 2018 AnandTech Storage Bench - Heavy
POST A COMMENT

43 Comments

View All Comments

  • DigitalFreak - Wednesday, February 6, 2019 - link

    Billy - how far are we away from having 4TB M.2 NVMe drives? Reply
  • Billy Tallis - Wednesday, February 6, 2019 - link

    Most SSD vendors could produce 4TB double-sided M.2 drives using off the shelf parts. Putting 4TB on a single-sided module would require either going with a DRAMless controller, stacking DRAM and the controller on the same package, or using 1Tb+ QLC dies instead of TLC.

    So currently any 4TB M.2 drive would have at least some significant downside that either compromises performance, restricts compatibility, or drives the price up well beyond twice that of a 2TB M.2 drive. There's simply not enough demand for such drives, and likely won't be anytime soon.
    Reply
  • Mikewind Dale - Wednesday, February 6, 2019 - link

    That drop in performance for a full drive in the Heavy - and even the Light!! - tests is worrying. They're right around the level of a SATA SSD.

    My question is, how full is full? If you fill the drive up 99%, is its performance closer to empty or full? With all my SSDs, I typically leave about 10% of the drive unallocated (unpartitioned). How would the drive perform in this state?

    I would be interested in seeing results for a drive that is almost full, but not quite full. I imagine that most people don't use their drives up until the final MB is used. Still, if a cost-conscious person is trying to get their money's worth, they might use the drive until it's 90-something percent full. Until recently, I was using a 512 GB SATA SSD with a real capacity of 476.8 GB. I used it until I was using 420 GB, at which point I upgraded to a 2 TB drive. So I was using 88% of its capacity. To me, that seems like a reasonable usage to test - not quite full, but almost full.
    Reply
  • Targon - Wednesday, February 6, 2019 - link

    I would suspect that the reason for this might be thermal throttle issues. Throw a heat sink on there, and the performance downgrade might disappear. The versions with a pre-installed heatsink might be worth the money, depending on how much it would cost to buy a SSD heatsink at this point(I haven't looked). Reply
  • BillyONeal - Wednesday, February 6, 2019 - link

    Seems more likely to be reduction in the size of the SLC cache -- see the the filling the drive tests where there are 3 distinct phases depending on how much space is actually in use. Reply
  • jabber - Thursday, February 7, 2019 - link

    I must admit I still leave a few GB spare/unallocated on any SSD I install. 2GB on a 120GB, 4GB on a 240GB and 8GB on a 500GB. Old habits. Reply
  • reactor_au - Thursday, June 13, 2019 - link

    I was wondering the same thing, how full can one get before performance drops off the cliff like in the benchmarks? Its a very import detail to omit! Reply
  • Luckz - Friday, November 29, 2019 - link

    At 80% full it was really tragic in this review of the 256GB size https://pclab.pl/art79361-9.html Reply
  • Mikewind Dale - Wednesday, February 6, 2019 - link

    I also notice that these drives don't have an active power state less than 3.8W. That's unfortunate, because as Ganesh T S noted in his Anandtech review of the MyDigitalSSD M2X M.2 NVMe SSD Enclosure, that enclosure will only work with SSDs that have an active power state less than 3.8W.

    I think this is important because it determines whether you can continue to use the SSD as a portable drive after you upgrade later. If you replace your 2 TB with a 4 or 8 TB SSD someday in the future, it will be nice to know that you can repurpose your 2 TB as an external drive.

    Also, it determines whether you can easily upgrade your SSD when all your M.2 slots are full. Whenever I upgrade a SATA boot drive, I typically use an external USB enclosure to clone the current SATA drive (still installed internally) to the new SATA drive (inside the enclosure). Then I can swap the two drives, and my computer will transparently use the new drive. With M.2, this is even more important because many motherboards have only two M.2 sockets. So if you have both M.2 sockets filled and try to upgrade one of the M.2 drives, you'll have a bit of a challenge. You could buy a PCIe-M.2 card and use that, but using an external USB enclosure is more convenient.

    So I'd like to see more M.2 drives with a sub-3.8 W active power state. The Samsung 970 EVO Plus has a 3.4 W active state, so it passes this test.
    Reply
  • MrSpadge - Wednesday, February 6, 2019 - link

    I love ADATA's naming scheme! It's so easily memorable and has more X's than any other brand. Reply

Log in

Don't have an account? Sign up now