The Intel Optane SSD 900P 280GB Review
by Billy Tallis on October 27, 2017 9:30 AM ESTConclusion
The Intel Optane SSD 900P is an amazing piece of technology, but one whose benefits are difficult to fully enjoy. Its 3D XPoint memory enables it to break almost all the performance records, but some where difference to Optane SSD performance is too small to justify paying more than twice the price per GB. However, there are some scenarios where the Optane SSD absolutely blows away the competition and justifies its premium.
The low queue depth random read performance of the Optane SSD is several times faster than any flash-based SSD has attained. Mixed workloads that include a substantial component of random read operations also perform quite well, and throwing some write operations into a stream of reads barely impacts the read performance.
The Optane SSD 900P has enough space to be used as general-purpose storage, and larger capacity models can be introduced as necessary. There's no need to tangle with Intel's caching software and its associated platform lock-in as with the Optane Memory modules.
The biggest problem the Optane SSD 900P faces today is that few desktop users have workloads that stress the storage system enough for the Optane SSD to shine. Mechanical hard drives have not disappeared from use as primary storage, and most software for desktop and workstation use is still designed with their performance limitations in mind. If budget SSDs had relegated hard drives to backup and archival use years ago, then the software landscape would probably be more ready to take advantage of the speed offered by the Optane SSD 900P. Instead, the Optane SSD as a consumer/prosumer product only makes sense in a few niches. Most users- even those with relatively intense storage performance needs - will be better served by high quality flash SSDs like the Samsung 960 PRO.
The price of the Intel Optane SSD 900P is accessible enough that many enthusiasts will pay the premium to have the bragging rights of the fastest SSD money can buy. Workstation users who have massive datasets that don't fit in RAM will jump at the chance to buy a faster scratch drive. And if that isn't enough to clear the shelves, then enterprise customers who need high performance but don't need the extreme write endurance of the Intel Optane SSD DC P4800X can get the 900P with its solid 10 DWPD endurance rating for a third the price per GB. The Optane SSD 900P isn't for everybody, but it will nonetheless be a successful product, and Intel won't have any trouble selling them.
The long-term prospects for Intel's Optane SSDs look pretty good, too. The pricing doesn't leave Samsung a lot of room to introduce a Z-NAND based consumer SSD. No other alternative nonvolatile memory technology is close to being ready to challenge 3D XPoint. Intel could improve the sequential transfer speeds, but they're good enough for now. They'll need to deliver a big jump in performance when they adopt PCIe Gen 4, but that shouldn't be a challenge: increasing the controller's channel count from 7 to the 12 channels used by their current NAND flash controller or the 18 used by their first NVMe controller will bring plenty of extra throughput. The only question will then be over the power consumption. The latency is already close to being as low as possible over a PCIe link, and NVDIMMs with 3D XPoint won't be making Optane SSDs obsolete in the consumer market anytime soon.
In many ways, the performance profile of the Optane SSD is far simpler than that of NAND flash based SSDs. The Optane SSD 900P performs just as well when it is full as when it is almost empty. It performs quite consistently over time, with far fewer high-latency outliers thanks to the lack of garbage collection stalls. Unlike with flash-based SSDs, it is not necessary to buy the largest model to get the highest possible performance; the 280GB model we tested should be very similar to the 480GB model (which we're waiting for review). It doesn't matter whether TRIM commands are used, and it's never necessary to perform a secure erase operation to restore degraded performance.
Intel has almost taken all the fun out of testing a SSD.
Facebook
Twitter
RSS
205 Comments
View All Comments
ddriver - Friday, October 27, 2017 - link
"MLC/TLC NAND treated like SLC"That sounds like "a snail treated as a cheetah". I bet feeding antelopes to a snail will make it as fast as a cheetah.
There is already a huge gap in access performance between MLC and TLC. TLC drives turn pathetic the moment they run out of cache. It is physically impossible to store multiple bits and access as if it is a single bit. Neither in terms of performance, nor in terms of endurance.
We haven't even seen what SLC is truly capable of, only the very early SSDs had SLC, and back then they were crippled by the primitive and tremendously under-powered controllers. At the medium level, SLC is insanely fast.
MFinn3333 - Friday, October 27, 2017 - link
OK, no that is simply not true.Fujtisu made a drive in 2014 entirely of Intel 25nm SLC and a Sandforce 2281 controller. It was called Fujitsu FSX 240GB (And 120GB). It kicked ass for it's class and time but it is nowhere and I mean nowhere near the speeds that are here or what you are claiming.
I set a couple of them up in RAIeD-0 through both hardware and software RAID and while it did often saturat the bandwidth but only with sequential transfers. The only number that it could ever come close to matching anything here is when you are talking random writes. I could easily hit 220MB/s at 4K Random Write but it's Random Read speed was around 50MB/s at the best of times.
SLC was and is awesome and I feel comfortable with doing horrible things to those drives (I defragged them for no reason, compiled code, did multiple virtual machines including Windows 3.11, and more benchmarking than any person ever should) but it's time is over.
Get over it.
ddriver - Friday, October 27, 2017 - link
Oh wow, you are some kind of a tech genius, expecting raid to boost something other than sequential access.2014 is ancient history in the world of SSD controllers. SLC on the physical level is capable of 500-600 times better performance than what the "best" SLC controller could squeeze out of it.
I am not saying this isn't the case for xpoint as well, I am just saying SLC is far more capable than what people can imagine.
MFinn3333 - Friday, October 27, 2017 - link
"We haven't even seen what SLC is truly capable of, only the very early SSDs had SLC, and back then they were crippled by the primitive and tremendously under-powered controllers. "That is what you wrote, not about SSD's a few years ago but the very early ones. You moved your own goalposts.
SLC is about 4x faster than MLC. You are claiming it to be 500x while showing little to no evidence and roasting Intel for their claims. So either put up or shut up with your evidence.
Reflex - Friday, October 27, 2017 - link
SLC does not obey physics. Gotta take that into account. ;)chrnochime - Wednesday, November 1, 2017 - link
No comment when the other guy mentions 2014 is not "very early SSD"? Come on now. Your prejudice against anything other than SLC is showing. We all know early SSDs go back much earlier than even 2007.edzieba - Saturday, October 28, 2017 - link
If SLC were truly better and cheaper than PCM, then companies would be using it and undercutting the competition with their cheaper, faster drives with lower production costs.jospoortvliet - Friday, November 3, 2017 - link
Indeed: https://xkcd.com/808/extide - Friday, October 27, 2017 - link
Regardless of the tests used here this thing is one of the fastest storage devices available PERIOD, and it's honestly priced pretty well. It's cheaper/GB than the first SSD I bought, in fact. I could see this being used for large swaps on servers that need a huge memory footprint for a lot cheaper than a shitload of RAM, or as a ZFS L2ARC or ZIL, or for hosting a ton of VM's or for running databases off of, etc.BUT you are ddriver, the king of cynicism, so I can at least say "I got what I expected."
ddriver - Friday, October 27, 2017 - link
I am also the king of "1000 times better means 1000 times better" ;)I too got what I expected, because I expected that "1000 times better" to be a lie.
If you look at my comments, I am actually 100% objective about acknowledging the benefits of hypetane. Which is where my true biggest fault lies. How dare I be objective rather than expressing nothing short of complete awe and admiration?