Burst IO Performance

Our burst IO tests operate at queue depth 1 and perform several short data transfers interspersed with idle time. The random read and write tests consist of 32 bursts of up to 64MB each. The sequential read and write tests use eight bursts of up to 128MB each. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

QD1 Burst IO Performance
Random Read Random Write
Sequential Read Sequential Write

The WD Black SN850 turns in excellent scores on almost all of the burst IO tests. For random reads, it edges out the Intel SSD 670p to set a new record for flash-based SSDs, and even when testing beyond the bounds of any possible SLC caching it is only 2% slower than the MLC-based Samsung 970 PRO. For random writes the WD Black SN850 is slightly slower than the Phison E16 drive, but otherwise is s clear step up in performance from the rest of the field. When testing sequential transfers on a small slice of the drive, the SN850 is substantially faster than everything else, but when testing across 80% of the drive its sequential read performance drops dramatically and is beat by the Samsung 980 PRO and several of the faster PCIe 3.0 drives.

Sustained IO Performance

Our sustained IO tests exercise a range of queue depths and transfer more data than the burst IO tests, but still have limits to keep the duration somewhat realistic. The primary scores we report are focused on the low queue depths that make up the bulk of consumer storage workloads. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

Sustained IO Performance
Random Read Throughput Power Efficiency
Random Write Throughput Power Efficiency
Sequential Read Throughput Power Efficiency
Sequential Write Throughput Power Efficiency

On the longer random read test, the WD Black SN850 doesn't quite stand out from the best performance offered by other drives with newer flash. But on the other three workloads the SN850 is clearly superior, with significant performance leads over the rest of the competition. Its power consumption is consistently on the high side and in some cases it is drawing more than any of the other drives, but the performance is high enough that the efficiency scores are all good.

Random Read
Random Write
Sequential Read
Sequential Write

For random reads, the SN850 eventually ramps up to around 4GB/s or 1M IOPS at the end of the test, which is significantly faster than any other drive that we've tested so far on this new test suite. However, when testing across 80% of the drive instead of just a 32GB slice, the random read performance falls to roughly the same level as the Samsung 980 PRO.

For random writes, the SN850's performance scales up a bit quicker than the 980 PRO, but it hits a throughput limit sooner and the 980 PRO ends up being much faster for random writes to the SLC cache at high queue depth.

For sequential reads, the SN850 ends up slightly faster than the 980 PRO, but when testing across 80% of the drive the Samsung reaches full performance with a lower queue depth. For sequential writes the SN850 is again a bit faster than the 980 PRO and this time it doesn't need higher queue depths to reach full speed, but it also starts running out of SLC cache before the test is over while the 980 PRO maintains full performance through the end of the test.

Random Read Latency

This test illustrates how drives with higher throughput don't always offer better IO latency and Quality of Service (QoS), and that latency often gets much worse when a drive is pushed to its limits. This test is more intense than real-world consumer workloads and the results can be a bit noisy, but large differences that show up clearly on a log scale plot are meaningful. For more details, please see the overview of our 2021 Consumer SSD Benchmark Suite.

The WD Black SN850 starts off this test with good random read latency, but around 80k IOPS it shifts gears and latency spikes alarmingly. It actually improves a few times later in the test so by the time the drive is approaching its throughput limit, it is only a bit slower than the Samsung 980 PRO.

Trace Tests: AnandTech Storage Bench and PCMark 10 Advanced Synthetic Tests: Block Sizes and Cache Size Effects
POST A COMMENT

83 Comments

View All Comments

  • Billy Tallis - Friday, March 26, 2021 - link

    A UPS vs power loss protection capacitors defend against slightly different sets of failure conditions, with a lot of overlap. A UPS will help save more data when the utility power goes out, but PLP caps will save data that a UPS couldn't if your PSU blows up or some other component failure inside the PC causes it to crash hard. Reply
  • wr3zzz - Friday, March 19, 2021 - link

    Is there a benchmark that shows "real-world' performance are not worth the premium between PCIe 3.0 and PCIe 4.0? Like how it translates in terms of time saved. CPU benchmark for web page loading is a good example. The benchmark numbers might be 2x or 10x but it doesn't mean anything if it means 1 second vs. 0.1 second. On the other hand, the difference from HDD to SSD is from going minutes to seconds.

    The delta between 4.0 and 3.0 is so wide now. It doesn't do us any good by showing benchmark numbers that are 2x across the board between SN850 and SN750 but then recommend the value proposition just isn't there for PCIe 4.0 in the real world.
    Reply
  • oRAirwolf - Friday, March 19, 2021 - link

    I do agree that it would be nice if Anandtech would add some real world benchmarks to show use scenarios for the most common tasks like loading Windows, various video game load times, common program launching, etc. I have a very high-end rig but I still boot off of 3 x Samsung 850 evos in RAID 0 because as far as I can tell, there is no significant benefit to switching to an mvne drive for my use case of working from my desktop providing high level tech support and gaming. Reply
  • Oxford Guy - Sunday, March 21, 2021 - link

    You should be able to compare the 'light' workload of 3.0 drives to the 'light' workload of 4.0 drives using their Bench comparison page.

    I'd pay particular attention to latency results. High latency can make things feel slow.

    'The benchmark numbers might be 2x or 10x but it doesn't mean anything if it means 1 second vs. 0.1 second.'

    Not true. .1 second latency (time to wake to do something) is a lot snappier than 1 second latency.
    Reply
  • Morawka - Friday, March 19, 2021 - link

    Excellent review. See if you guys can a Sabrent Rocket 4 Plus sample and do a review on it! It's supposedly a little faster that the WD 850 in synthetics, but perhaps a little slower in real-world usage. It will be a close match up, that's for sure. Reply
  • Billy Tallis - Friday, March 19, 2021 - link

    I don't have a Rocket 4 Plus, but I have Micro Center's equivalent: Inland Performance Plus. It's currently running The Destroyer, and has already completed the synthetic tests. https://www.anandtech.com/bench/product/2732?vs=27... Reply
  • Oxford Guy - Sunday, March 21, 2021 - link

    Glad to hear there is a firmware update. I wasn't sure that Inland drives would be eligible for them. Reply
  • Spunjji - Friday, March 19, 2021 - link

    "MLC is now dead, and there's no compelling reason to bring it back"

    Oooof, that's gonna chafe the NAND conspiracy-theorists who frequent these hallowed comments 😅
    Reply
  • Oxford Guy - Sunday, March 21, 2021 - link

    3D fabrication made TLC viable. It has not remedied the deficiencies of QLC and now Intel is reportedly planning to push PLC onto consumers.

    There is nothing theoretical about:

    • The fact that QLC only offers 30% more density, despite having a lot more voltage states than a 30% increase (diminished returns)

    • QLC drives have not been priced low enough to make them worthwhile

    • PLC is going to be worse

    • Economy of scale is actively working against consumer value, by inflating the price of TLC

    Failure of trolling noted.
    Reply
  • FunBunny2 - Sunday, March 21, 2021 - link

    "There is nothing theoretical about:"

    the fact that, what, no more than a handful make NAND. if they choose, whether through collusion or simultaneous profit-seeking, to sell only QLC and PLC what are you going to do about it? invoke the Defense Production Act to force SLC production at reduced prices? unfettered capitalism never favours the consumer.
    Reply

Log in

Don't have an account? Sign up now