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

  • eddieobscurant - Wednesday, February 6, 2019 - link

    Nice review , as always although I disagree with your conclusion. Peak performance is what most people want. Reply
  • Billy Tallis - Wednesday, February 6, 2019 - link

    My reviews are intended to advise consumers who are buying SSDs to increase their productivity, not people who are trying to set a high score on Crystal Disk Mark.

    People who care about real-world productivity rather than CDM scores should recognize that imperceptible improvements to peak performance are probably not worth the sacrifice of significant regressions in performance on niche heavy workloads. For a lot of users, both SM2262 and SM2262EN drives are fast enough. Beyond those lighter use cases, I think it will be more common to find the SM2262EN coming up short in a meaningful way than to find it providing a tangible performance advantage over SM2262.
    Reply
  • mapesdhs - Thursday, February 7, 2019 - link

    I can't help wondering how some of the old favourites would behave in these comparisons, the 950 EVO/Pro, 960s, etc. Have things really moved on that much? Reply
  • Billy Tallis - Thursday, February 7, 2019 - link

    We have at least partial test results in Bench for most of the old drives that aren't worth including in every review: https://www.anandtech.com/bench/product/2219?vs=23... Reply
  • eddieobscurant - Thursday, February 7, 2019 - link

    It's not about crystal disk mark score. It's about almost no one of the everyday user, playing games, surfing the web and using microsoft office, will come near your "light" test, let alone "heavy" or "torture".

    Most of them need high random reads for their computer to feel snappy and responsive, and a big enough a slc cache to accommodate a full bluray of writes.
    Reply
  • Billy Tallis - Thursday, February 7, 2019 - link

    "Most of them need high random reads for their computer to feel snappy and responsive,"

    They already have that. Further increases to random read performance won't make the system feel any more responsive during light workloads, as demonstrated by SYSmark. High-end NVMe SSDs are already way past the point of diminishing returns for peak random read speeds, especially for lighter workloads where a few GB of DRAM used by the OS for caching is enough to almost completely decouple storage performance from application responsiveness.
    Reply
  • eddieobscurant - Friday, February 8, 2019 - link

    So, you're saying that optane doesn't feel more responsive to you, or that the high random reads of optane isn't responsible for feeling more responsive than a high end nvme ssd ? Reply
  • Dark_wizzie - Wednesday, February 6, 2019 - link

    Why does perf drop on 2tb model? Reply
  • Dark_wizzie - Thursday, February 7, 2019 - link

    For low qd random reads, sorry. Reply
  • Dark_wizzie - Thursday, February 7, 2019 - link

    And... serves me right for commenting before finishing the last page of the article. >.>
    oh well.
    Reply

Log in

Don't have an account? Sign up now