AnandTech Storage Bench - Heavy

Our Heavy storage benchmark is proportionally more write-heavy than The Destroyer, but much shorter overall. The total writes in the Heavy test aren't enough to fill the drive, so performance never drops down to steady state. This test is far more representative of a power user's day to day usage, and is heavily influenced by the drive's peak performance. The Heavy workload test details can be found here. This test is run twice, once on a freshly erased drive and once after filling the drive with sequential writes.

ATSB - Heavy (Data Rate)

The Kingston A1000's average data rates on the Heavy test make it clearly faster than most SATA drives, but not by a huge amount. The 960GB model seems to perform a bit better when full, but when the test is run on an empty drive both capacities offer about the same level of performance.

ATSB - Heavy (Average Latency)ATSB - Heavy (99th Percentile Latency)

The average and 99th percentile latencies from the Kingston A1000 on the Heavy test are a bit better than the scores from the MyDigitalSSD SBX, especially when the test is run on a full drive. This shows that the larger spare area reserved by the A1000 does have some impact. The 99th percentile latencies are not much better than SATA drives, but the average latencies from the A1000 do show a clear advantage over the Crucial MX500.

ATSB - Heavy (Average Read Latency)ATSB - Heavy (Average Write Latency)

The Toshiba RC100 showed that Toshiba's BiCS3 3D TLC can offer great read latency, but the Phison E8-based A1000 and SBX need to improve quite a bit to match that standard. For average write latency on the Heavy test, the A1000 outperforms most other low-end NVMe SSDs and SATA drives, but can't match high-end NVMe drives.

ATSB - Heavy (99th Percentile Read Latency)ATSB - Heavy (99th Percentile Write Latency)

The 960GB Kingston A1000 benefits greatly from its large spare area, with 99th percentile read and write latencies that are almost the same whether the Heavy test is run on  a full drive or an empty drive. The 480GB A1000 has a small QoS advantage over the 512GB SBX for the full-drive case. The read latency of the Toshiba RC100 again stands out as surprisingly good even though its 99th percentile write latency is nothing special. Compared to the Crucial MX500 SATA SSD, the A1000 and similar low-end NVMe SSDs only improve on 99th percentile read latencies, with 99th percentile write latencies showing little or no improvement.

ATSB - Heavy (Power)

On the Heavy test, the Kingston A1000 again has a bit worse power efficiency than the MyDigitalSSD SBX, which also makes it clearly more power-hungry than SATA SSDs but still more efficient than most NVMe SSDs.

AnandTech Storage Bench - The Destroyer AnandTech Storage Bench - Light
Comments Locked

26 Comments

View All Comments

  • romrunning - Monday, July 2, 2018 - link

    Any NVMe drive that doesn't beat the Intel 600p in every category shouldn't be made. I really wish all mfgs would keep NVMe drives to at least PCI x4 & 8 channels with a minimum performance level that is much higher than SATA. The bar for the next tech level of storage products should be at a higher level than the previous.

    Forget the low-end marked for NVMe. SATA can easily take care of any needs there.
  • peevee - Monday, July 2, 2018 - link

    Given that only Samsung barely saturates PCIex2, and only on artificial tests, at this point even x4 is useless, let alone x8. They'll need many more channels.

    Actually, I'd prefer x1 for ultra-low-power, if it is fully saturated in more or less real-life tests (like AT's "Light"), vs x4 which only saturates 1/10th of x1 capacity.
  • romrunning - Tuesday, July 3, 2018 - link

    I was thinking more along the lines of every mfg creating NVMe drives have to attain to a minimum performance level. It becomes easier to understand what is higher-performing from the end-user's perspective. So if NVMe's minimum performance level is 2x SATA, then anytime you see NVMe you know it's better than SATA. Too bad whatever storage consortium finalized specs for NVMe didn't require min perf levels for storage.

    It's annoy to me when these mfgs put out "new" drives that don't exceed the older tech.
  • FunBunny2 - Monday, July 2, 2018 - link

    " I really wish all mfgs would keep NVMe drives to at least PCI x4 & 8 channels with a minimum performance level that is much higher than SATA. "

    some wag put it, "you sell the sizzle, not the steak".
  • Gunbuster - Tuesday, July 3, 2018 - link

    Okay Kingston, reviews are done, feel free to swap in cheaper/slower NAND chips. ;)
  • Ratman6161 - Tuesday, July 3, 2018 - link

    In the ATSB Heavy Data Rate chart, for the 1TB 970 EVO, I think you have the full and empty numbers transposed. I.e. you show 525 empty and 635 full. I assume that should be 635 empty and 525 full?
  • Billy Tallis - Tuesday, July 3, 2018 - link

    They're not transposed. I'm not sure what happened with those test runs, but I'm re-running them. I do know that Samsung drives lie about when they've finished a secure erase, so it's possible the "empty" drive test run was still working on an erase operation in the background even though I try to ensure all drives have plenty of idle time to finish cleaning up after they claim to be done erasing.
  • leexgx - Wednesday, July 4, 2018 - link

    Surprised it works like that when using secure erase, zapping page area and all NAND chips should not take long to do, also what can i use to use secure erase on all drives (seagate own tool seems to lack it, it has full erase but it's not secure erase and it killed my seagate firecuda doing it)
  • SanX - Tuesday, July 3, 2018 - link

    Everyone here knows that in the shops the average Joe will see on the product tag "Sequential Read 1500 MB/s" which is plain lie and conveniently keeps mum about this. Which test gives 1500, show me? At best 2-3 times less.

    This site degraded long ago to serve salespeople.
  • rocky12345 - Wednesday, July 4, 2018 - link

    Great review but this drive does seem rather unimpressive for sure. It acts like it doe snot have a dram cache at all since most if it's scores are well below the mark. I like Kingston for their memory products which work well in the systems I build for my clients. These NVMe drives that are considered lower end give a false picture of great speed and performance because NVMe drives are known for their great performance level. Then you get these drives trying to break into this sector and do not perform any where close to what you would expect from a NVMe drive. Heck my samsung 860 Pro 512GB Sata SSD can get better numbers in a lot of the tests done in the review than these cheap low end NVMe drives and it is only based off of the Sata port and limited to a max 600MB's from the port itself that is kinda sad if you think about it really.

Log in

Don't have an account? Sign up now