Caching And Tiering: Intel Optane Memory H20 and Enmotus FuzeDrive SSD Reviewed
by Billy Tallis on May 18, 2021 2:00 PM EST- Posted in
- SSDs
- Storage
- Intel
- SSD Caching
- 3D XPoint
- Optane
- Optane Memory
- Tiger Lake
Our suite of Linux-based synthetic tests cannot directly measure how the Intel Optane Memory H20 or Enmotus FuzeDrive SSD behave when used with their respective caching or tiering software, but we can investigate how the individual components of those storage systems perform in isolation. The tests on this page are all conducted on our usual AMD Ryzen testbed.
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.
 |
|||||||||
| Random Read | Random Write | ||||||||
| Sequential Read | Sequential Write | ||||||||
The Intel Optane cache devices still have a huge lead over other storage technologies for random read performance, but the most notable improvements the Optane Memory H20 makes over the H10 are for random and sequential writes, where the little Optane devices most drastically underperform NAND SSDs. The small 32GB cache is now almost as fast as the 118GB Optane 800P was, but that's still not enough to match the performance of a decent-sized NAND SSD.
The SLC portion of the Enmotus FuzeDrive SSD doesn't particularly stand out, since these burst IO tests are mostly operating out of the SLC caches even on the regular NAND SSDs. However, the static SLC does clearly retain full performance on these tests even when full, in stark contrast to most of the drives that rely on drive-managed SLC caching.
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.
 |
|||||||||
| Random Read | Throughput | Power | Efficiency | ||||||
| Random Write | Throughput | Power | Efficiency | ||||||
| Sequential Read | Throughput | Power | Efficiency | ||||||
| Sequential Write | Throughput | Power | Efficiency | ||||||
There are no big changes to note for the Optane cache device on these longer-running tests that bring in some higher queue depths. The main advantages of Optane devices are at low queue depths, especially the QD1 that was already covered by the burst IO tests. The SLC portion of the Enmotus FuzeDrive does start to show some performance loss on the write tests when it is mostly-full, illustrating that SLC NAND is not immune to the performance impacts of background garbage collection with time-consuming block erase operations.
 |
|||||||||
| Random Read | |||||||||
| Random Write | |||||||||
| Sequential Read | |||||||||
| Sequential Write | |||||||||
Digging deeper into the performance results, the Optane portion of the H20 shows the results we expect, reaching its full performance at fairly low queue depths and showing high consistency.
The SLC portion of the Enmotus FuzeDrive SSD shows many of the characteristics we're used to seeing from NAND-based SSDs, including the common Phison trait of poor sequential read performance at low queue depths.
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 Optane cache on the H20 can't beat the maximum random read throughput available from a mainstream or high-end TLC SSD, but until the Optane hits its throughput limit it has unbeatable latency. The SLC portion of the Enmotus FuzeDrive SSD not really any faster on this test than a good TLC drive.
Facebook
Twitter
RSS
45 Comments
View All Comments
haukionkannel - Wednesday, May 19, 2021 - link
All ”highend” ssd are soon gonna be qlc and middle and low range will go plc…So just wait the things to get even worse!
;)
edzieba - Friday, May 21, 2021 - link
Ah, the QLC brigade is here, with the same Dire Warnings Of Horrible Doom that previously fell flat for MLC and TLC, but THIS time will totally come true (or we'll cross out the Q and put P and protest against the evils of PLC next year!).kepstin - Tuesday, May 18, 2021 - link
If you were to somehow get one of these Intel drives and plug it into an unsupported system, will it just show up as 2 separate NVMe drives? Would you be able to use it with hardware agnostic caching solutions like PrimoCache on Windows or bcache/dm-cache on Linux?drexnx - Tuesday, May 18, 2021 - link
sounds like the host system just sees it as a 32gb optane SSDBilly Tallis - Tuesday, May 18, 2021 - link
Depends on what the host system is, and what kind of slot. Only the supported Intel systems can initialize PCIe links to both sides. For the H10 review, I made a chart of all the systems I'd tried: https://www.anandtech.com/show/14249/the-intel-opt...If the slot is only PCIe x1 or x2, you get the NAND. If it's x4, you might get the NAND or you might get the 3DXP.
kepstin - Tuesday, May 18, 2021 - link
Ah, so there's no PCIe bridge/switch on the device itself? I guess they're relying on the upstream bridge of the M.2 slot supporting bifurcating the 4× link into 2×/2×.Billy Tallis - Tuesday, May 18, 2021 - link
Correct. The H10 and H20 rely on upstream port bifurcation support. I think there's also a proprietary element to it, but bifurcation down to x2 links is less widely supported than bifurcation down to x4 links anyways.A PCIe switch would have been nice, but wouldn't fit. And this product line isn't important enough for Intel to make a big new custom ASIC for, either a SSD controller that can speak to both 3DXP and QLC, or adding PCIe switch/passthrough support to one of the two controllers.
Kurosaki - Tuesday, May 18, 2021 - link
Maybe next gen then?...Tomatotech - Tuesday, May 18, 2021 - link
The H20 is just a placeholder while the Optane team treads water and begs the rest of Intel to let them release a proper drive. This H20 looks like a greyhound with bricks tied to its neck. Absolutely lovely latency and random 4K performance would be a credit to any high-end workstation. But it's crippled by a shit implementation.Apple was doing tiered drive storage nearly 10 years ago with their Fusion drives, and as the Enmotus tiered drive shows, it can do amazing things. This is how the H20 should be set up.
I say the H20 is treading water; it should be on PCIe 4.0 but because of Intel's shenanigans with PCIe 4.0 the Optane team are crippled and can't release this drive with the backing support it needs. Hopefully the next model, maybe the H30, will have PCIe 4.0 and then it'll finally be a decent overall drive.
Probably not, given the sad history of Intel shooting Optane in the foot. They could have released this drive several years ago, and it would have been excellent then, but normal NAND drives are improving all the time and it's just too little, too late.
Spunjji - Wednesday, May 19, 2021 - link
"a greyhound with bricks tied to its neck" - succinct. 👍