Whole-Drive Fill

This test starts with a freshly-erased drive and fills it with 128kB sequential writes at queue depth 32, recording the write speed for each 1GB segment. This test is not representative of any ordinary client/consumer usage pattern, but it does allow us to observe transitions in the drive's behavior as it fills up. This can allow us to estimate the size of any SLC write cache, and get a sense for how much performance remains on the rare occasions where real-world usage keeps writing data after filling the cache.

The Sabrent Rocket Q takes the strategy of providing the largest practical SLC cache size, which in this case is a whopping 2TB. The Samsung 870 QVO takes the opposite (and less common for QLC drives) approach of limiting the SLC cache to just 78GB, the same as on the 2TB and 4TB models.

Sustained 128kB Sequential Write (Power Efficiency)
Average Throughput for last 16 GB Overall Average Throughput

Both drives maintain fairly steady write performance after their caches run out, but the Sabrent Rocket Q's post-cache write speed is twice as high. The post-cache write speed of the Rocket Q is still a bit slower than a TLC SATA drive, and is just a fraction of what's typical for TLC NVMe SSDs.

On paper, Samsung's 92L QLC is capable of a program throughput of 18MB/s per die, and the 8TB 870 QVO has 64 of those dies, for an aggregate theoretical write throughput of over 1GB/s. SLC caching can account for some of the performance loss, but the lack of performance scaling beyond the 2TB model is a controller limitation rather than a NAND limitation. The Rocket Q is affected by a similar limitation, but also benefits from QLC NAND with a considerably higher program throughput of 30MB/s per die.

Working Set Size

Most mainstream SSDs have enough DRAM to store the entire mapping table that translates logical block addresses into physical flash memory addresses. DRAMless drives only have small buffers to cache a portion of this mapping information. Some NVMe SSDs support the Host Memory Buffer feature and can borrow a piece of the host system's DRAM for this cache rather needing lots of on-controller memory.

When accessing a logical block whose mapping is not cached, the drive needs to read the mapping from the full table stored on the flash memory before it can read the user data stored at that logical block. This adds extra latency to read operations and in the worst case may double random read latency.

We can see the effects of the size of any mapping buffer by performing random reads from different sized portions of the drive. When performing random reads from a small slice of the drive, we expect the mappings to all fit in the cache, and when performing random reads from the entire drive, we expect mostly cache misses.

When performing this test on mainstream drives with a full-sized DRAM cache, we expect performance to be generally constant regardless of the working set size, or for performance to drop only slightly as the working set size increases.

The Sabrent Rocket Q's random read performance is unusually unsteady at small working set sizes, but levels out at a bit over 8k IOPS for working set sizes of at least 16GB. Reads scattered across the entire drive do show a substantial drop in performance, due to the limited size of the DRAM buffer on this drive.

The Samsung drive has the full 8GB of DRAM and can keep the entire drive's address mapping mapping table in RAM, so its random read performance does not vary with working set size. However, it's clearly slower than the smaller capacities of the 870 QVO; there's some extra overhead in connecting this much flash to a 4-channel controller.

Introduction AnandTech Storage Bench
Comments Locked

150 Comments

View All Comments

  • Silver5urfer - Friday, December 4, 2020 - link

    $900+Tax for this garbage QLC ? Utter shame. SSDs are most useful for gaming, better have 2x 860 PRO 4TBs even if it's expensive as it's going to be last one with MLC technology. And Movies, are fine with the HDD drives.

    High Capacity drives are mostly for the 4K, Movies and Anime etc content or maybe photos. Any NAS grade drive for the archiving purpose is much much better only issue is noise and size, best is to get the mechanical drives for high capacity.

    So this means consumers will get the shitty QLC technology for high capacity HDD, probably we will get Six Level Bullshit next time to even push more TBs. What a load of garbage.
  • Silver5urfer - Friday, December 4, 2020 - link

    1800 Sabrent QLC NVMe, 2880 Samsung QLC SATA

    Vs

    860 PRO 4TB 4800. That's all I need to know, 2x the storage and latest manufacturing processes and almost like 2 years after the 860Pro products and we have 1/4 for NVMe and 1/2 for SATA with 2x Storage, if 860PRO had 8TB it would be probably 8000+TBW endurance.
  • Oxford Guy - Friday, December 4, 2020 - link

    "latest manufacturing processes"

    With NAND that can be a bad thing, unless you're a manufacturer, that is.
  • Spunjji - Monday, December 7, 2020 - link

    ...and it would cost nearly twice as much money (assuming they could actually make the components fit in that 2.5" shell). You pay your money and take your choice.
  • Oxford Guy - Monday, December 7, 2020 - link

    QLC offers 30% more density than TLC.
  • Beaver M. - Saturday, December 5, 2020 - link

    I completely agree on the price. Its a joke. Far too close to TLC drives, in fact very comparable.

    But for games you will want NVMe drives from now on, because of DirectStorage.
  • Deicidium369 - Saturday, December 5, 2020 - link

    Which is not even available yet - but yeah RTX IO/MS DirectStorage will require an NVMe drive - I have a couple 980 Pro, but waiting on the new Phison based drives with 7GB/s R&W.
  • Beaver M. - Saturday, December 5, 2020 - link

    Yeah well, you dont buy SSDs to replace them just a few months later.
    Its stupid to not plan ahead when putting together a PC, especially nowadays when most PC components can easily be good enough for 5+ years.
  • Deicidium369 - Sunday, December 6, 2020 - link

    I build for 2 years - my current i9900K / 3090 (replacing the dual 2080TIs was one of the rare upgrades) is already over 2 years old - the 3090 is recent of course. When Rocket Lake drops - I will build the replacement for my current system - and by then the new Phisons will be available. I seem to remember it being released this month - but could be wrong.
  • Silver5urfer - Saturday, December 5, 2020 - link

    I don't really know how DirectStorage is going to impact performance. It will just enable higher and faster load times, It's already very fast on a PC due to dedicated VRAM and DRAM on top. I don't believe this RTX IO or whatever are going to have a massive impact. I have lots of TBs of backlog of old games which I love to play rather than new politically driven garbage with MTX and other crap and GaaS. Nothing much to lose tbh I will rather buy an 860 PRO x2 and install all the games on it and enjoy and get a WD RED / GOLD or Seagate Exos for the 4K and Movies etc.

Log in

Don't have an account? Sign up now