Intel Launches Mainstream Optane SSD 800P Series Based On 3D Xpoint Memory (hothardware.com) 36
MojoKid writes: Intel just launched a new family of consumer-targeted Optane solid state drives today, dubbed the Intel Optane SSD 800P. Unlike Intel Optane Memory sticks, which accelerate hybrid storage configurations with hard drives through intelligent data caching, or Intel's flagship Optane SSD 900P that's aimed squarely at hardcore enthusiasts with big budgets, these M.2 form factor Intel Optane 800P SSDs target the meat of the mobile and desktop markets, with higher capacities than Optane Memory but more affordable pricing than the 900P. In the benchmarks, the Optane SSD 800P series drives offered a mixed-bag of performance, with sequential transfers that top out at about 1.4GB/s, but with small file transfers, 4K random and mixed workloads, latency, and overall QoS looking strong. Intel will initially be offering two drives in the Optane SSD 800P series, with M.2 80mm 58GB and 118GB models. Suggested pricing for the drives is $129 for the 58GB capacity and $199 for the 118GB drive.
Trash (Score:1)
In every spec that matters, these are far worse than Samsung's 950 and 960 offerings.
Maybe next time, Intel!
Re: (Score:1)
"What matters" varies from person to person. These things have impressive endurance for their sizes and consistent low latency. There are uses for that.
I don't need them, but I'm sure someone else does. Let them buy it and go grab a 950 for yourself.
Re: (Score:2)
These things CLAIM to have impressive endurance. That remains to be seen.
During Xpoint's development, just about every promised spec, including endurance, was decreased by several orders of magnitude.
When the first generation of Xpoint devices launched, they included a huge amount of extra hardware for over provisioning. If Xpoint doesn't need it, why include it? And why so much of it (more than an SSD needs).
Further, the only endurance spec that matters is the one on the box. Samsung's products carry a
Re: Trash (Score:5, Informative)
Careful there, not sure about these particular ones but Intel's current crop emaciates the Samsung even though on paper and synthetical benchmarks (IOPS and transfer rate) the Samsung does better.
Testing it myself, the Samsung does good until you transfer ~2-3GB and then it drops like a brick to the lower 1000s of IOPS instead of the 100,000 or more it gave you.
The problem there is that Samsung gives you a good RAM cache (backed up with huge capacitors on their DataCenter models) but once you request synced rates or exceed that cache, the controller lags behind. In the mean time, the Intel continues chugging along at 70-90k IOPS.
Re: (Score:2)
You don't mean 'decimate' either, as that simply reduces by 10%. I think he meant 'annihilate'.
Re: (Score:2)
Careful there, not sure about these particular ones but Intel's current crop emaciates the Samsung even though on paper and synthetical benchmarks (IOPS and transfer rate) the Samsung does better.
Testing it myself, the Samsung does good until you transfer ~2-3GB and then it drops like a brick to the lower 1000s of IOPS instead of the 100,000 or more it gave you.
The problem there is that Samsung gives you a good RAM cache (backed up with huge capacitors on their DataCenter models) but once you request synced rates or exceed that cache, the controller lags behind. In the mean time, the Intel continues chugging along at 70-90k IOPS.
What drive are you testing, exactly? Is it an NVMe drive?
Are you using the proper drivers from Samsung? If you're on Windows, the default driver from MS is kind of shitty. You'll want to download the NVMe driver from Samsung. I'm not sure what the situation is on Linux.
The 960 series from Samsung has much better sustained IOPS performance after reaching steady state and in an extended load than the 950 series. It approaches that of the Samsung/Intel enterprise drives (which have gobs of over provisioni
Re:Trash (Score:5, Informative)
Relative latencies:
SRAM 1X
DRAM 10X
Optane 100X
NAND 100,000X
Rust 10,000,000X
https://hothardware.com/review... [hothardware.com]
The same gains as going from HDD to SDD (1000X) are realized again going from SDD to Optane.
Probably one of those things, you don't know why you'd even need it, until you have it, then you won't want to live without it.
Or you're not demanding enough to even notice either way.
Re: (Score:2)
Probably one of those things, you don't know why you'd even need it, until you have it, then you won't want to live without it. Or you're not demanding enough to even notice either way.
Or it's just not convenient to split the bits that could use Optane. I had the same issue with SSD/HDD, apart from bulk media (photos, audio, video) the primary space consumer is games. And a lot of that is cut scenes and textures and music loops and voice acting etc. that don't benefit a lot, while the main executable, scripts, maps, UI elements etc. is 1-5% of the game. But as long as Steam doesn't give me an easy way to say this 1GB go on SSD, the remaining 19GB go on HDD it's not worth the hassle, let'
Re: (Score:2)
I would be happy with 64-96 gigs of Optane ram, and no SSD
Probably be faster than my 80 gig X25 8 gig setup now.
Re: (Score:2)
If you care about latency you're going to want to just keep your shit in RAM. That's the trend on the enterprise side.
Consumers and even high end gamers do not benefit from that small latency advantage. Optane drives only beat traditional drives when you do synthetic, random access tests or crank the queue depth down for no reason. People moving large files don't benefit. Gamers don't benefit - assets are loaded from sequential blobs and (ideally) kept in RAM/VRAM. People working with media and shit do
Re: (Score:2)
I would say for every spec that matters, optane is a huge improvement over both the Samsung 950 and 960. I had two Samsung 950 PRO's (in RAID-0), and then replaced them with a single 960 PRO. Just like the linked article shows, performance on these are all over the place, often not demonstrating good performance until you hit high queue depths, which isn't "common" for a single user desktop.
Just a quick test I ran just now shows read speeds of 1903MB/s, and write of 1076MB/s. And then a second run show 1
What is Optane for? (Score:2)
Maybe it's just me, but I struggle to see the point of Optane as compared to a regular flash-based SSD.
From what. I can see, it's optimised as a high speed but small SSD that can then be used as a cache for a spinning HDD.
In the benchmarks I've seen however, it doesn't seem to be markedly faster than a fast M.2 NVMe SSD.
Re: (Score:3)
Maybe it's just me, but I struggle to see the point of Optane as compared to a regular flash-based SSD.
From what. I can see, it's optimised as a high speed but small SSD that can then be used as a cache for a spinning HDD.
In the benchmarks I've seen however, it doesn't seem to be markedly faster than a fast M.2 NVMe SSD.
It is on the wrong interface speaking the wrong protocol. It needs to be connected to the DDR interface for the low latency, fast writes, and word-unit access of 3D-Xpoint to shine. But we are still waiting for that product. When the first NVMe Optane came out, many watchers thought that it indicated some yield issues that more complex and higher latency NVMe interface could cover up but that would make it difficult to make a DDR connected device functional. Maybe Intel is still having trouble.
Re: (Score:3)
One thing causing problems is interrupt driven disk drivers instead of polling SSDs these days are so fast that polling is starting to make more sense than assuming it will take awhile for the disk to get back to you with the data.... there
Re:What is Optane for? (Score:5, Informative)
Well, NVMe is a multi-queue spec. The best drivers and chipsets for it will assign a command and response queue to each cpu in the system. This allows for both lockless queuing operation as well as polling with no cross-cpu contamination. In this regard, NVMe is far, far superior to AHCI (aka SATA, which has only one queue for multiple targets) and SAS chipsets (which typically are not multi-queue).
At 10uS, though, interrupt overhead (with MSI-X vectoring per-cpu) still yields superior cpu-v-data performance. Interrupt overhead is only around ~1uS or so. Still, its getting close. At lower latencies polling will definitely be a win. But even at 10uS, interrupt driven operation still leaves the cpu with extra clocks to do other work in that it wouldn't have with polling.
Another problem is that NVMe chipsets generally don't have anywhere near the 1023+ queues supported by the spec. They usually come in at no more than 31 queues, which is not enough to assign one to each cpu thread on heftier systems. The chipset spec can support a lot more... in fact, many more MSI-X interrupts can be supported as well, but we just don't see it out in the field.
Most chipsets only offer 8 queues, which is near worthless on modern multi-core cpus.
nvme0: Model SAMSUNG_MZVPV128HDGM-00000 BaseSerial S1XVNYAGA03031 nscount=1
nvme0: Request 64/32 queues, Returns 8/8 queues, rw-sep map (8, 8)
nvme1: Model Samsung_SSD_960_EVO_250GB BaseSerial S3ESNX0J219064Y nscount=1
nvme1: Request 64/32 queues, Returns 8/8 queues, rw-sep map (8, 8)
nvme2: Model INTEL_SSDPEKKW256G7 BaseSerial BTPY64430Q5B256D nscount=1
nvme2: Request 64/32 queues, Returns 8/8 queues, rw-sep map (8, 8)
nvme3: Model TOSHIBA-RD400 BaseSerial Z6TS10AUTPEV nscount=1
nvme3: Request 64/32 queues, Returns 7/7 queues, rw-sep map (7, 7)
nvme4: Model WDC_WDS256G1X0C-00ENX0 BaseSerial 170369420988 nscount=1
nvme4: Request 64/32 queues, Returns 16/16 queues, rw-sep map (16, 16)
nvme5: Model BPX BaseSerial 8B7107720F0823024374 nscount=1
nvme5: Request 64/32 queues, Returns 7/7 queues, rw-sep map (7, 7)
nvme6: Model PLEXTOR_PX-256M8PeG BaseSerial P02652102851 nscount=1
nvme6: Request 64/32 queues, Returns 16/16 queues, rw-sep map (16, 16)
Eventually we'll start to see chipsets that implement closer to the queue limit in the spec, at which point we can theoretically assign a queue pair to every active user thread using the storage. But for now I would be happy if chipsets just gave us enough queues to implement two per cpu thread (for priority separation).
Also, Intel NVMe SSDs are *NORTORIOUSLY* bad in multi-queue configurations. Performance is far poorer than other vendors placed in the same configuration. I think this is rather ironic, actually. Intel markets low latency, but their chipsets can't handle it in the real-life configurations that NVMe was designed for.
-Matt
Re: (Score:3)
Matt, thank you as always for your comments, and contributions.
If the Optane drive is significantly faster at low queue depths, does that mean that it is capable of maintaining a lower queue depth for longer by servicing them faster, preventing the depth from growing?
Expensive (Score:1)
Re: (Score:2)
Is this finally the fix? (Score:3)
Re: (Score:2)
Uh... You do realize that Micron is a US company, not asian, right?
Not really a consumer product (Score:5, Insightful)
These are not really consumer products. Basically what you get out of an Optane drive is more durability (hence 10DWPB instead of 0.3DWPD @ 5 year warranty), and low latencies at low queue depths ( 10uS @ QD1 instead of 30uS+ @ QD1 for a NAND drive, random read).
But that's it. Everything else about Optane is non-competitive with NAND, at least so far. The price is ridiculous, the throughput at higher queue depths isn't really all that impressive.
No consumer is going to notice the lower latencies at low queue depths for the types of activities Intel advertises the product for (such as gaming), because all of those activities involve bulk reading and writing which NAND does very well, and most involve a certain degree of sequential reading or writing which modern NAND drives (such as the Samsungs) optimize very well. At higher queue depths the Intel advantage goes away entirely, so it wouldn't move the needle even for concurrent random server workloads.
Consumers for the most part never hit the actual durability limits of a NAND drive. For one, even with the lower durability the NAND drive is typically going to be double or triple the capacity of the Optane drive at the same price point, and for two, consumer use cases do not usually do 10 full drive writes per day over the life of the device or anything even close to that.
Basically, like the idiotic optane 'disk cache' Intel tried to hawk last year, this drive is a pretty bad fit as a consumer device. In this offering Intel at least put the proper durability that Optane is *supposed* to have in the specs. Around 8900TB... nothing to sneeze at when most NAND drives have durabilities in the 200-400TB range. There is something to be said for that, even without real-life integrity/retention data available yet. But... it's still just not a consumer-oriented device.
-Matt
Re: (Score:2)
In this offering Intel at least put the proper durability that Optane is *supposed* to have in the specs. Around 8900TB... nothing to sneeze at when most NAND drives have durabilities in the 200-400TB range.
But the NAND drive has such relatively low endurance due to economic motivations. SLC NAND has significantly higher endurance at a higher price point. The consumer market requires the lower price point. What would the Optane endurance be when scaled and designed for the consumer-viable price point?
Re: (Score:2)
I just can't get over the capacities. Is the difference in performance between Optane and SSD significant enough to make the resulting complex architectures of including relatively small Optane drives into a storage environment even worth the effort?
My long-term take is that the best performing caching and tiering storage systems really only provided marginal performance improvements at significant price premiums and involved complex controller architectures to manage it. And of course the performance imp
2 orders of magnitude shy of belt and suspenders (Score:2)
Google calc: 10 * 58 GB / (3 GB/s * 24 hours ) = 0.22% gum-stick NVME duty cycle
I've plucked interface bandwidth from the article:
With a mere two more orders of magnitude of endurance, this small Optane product w
What? (Score:1)
I thinking I'm missing the point of these.
More expensive than a Samsung EVO 960.
Smaller than a 960.
Half the read performance of a 960.
Half the write performance of a 960.
I was excited when I first read the headline, I thought, FINALLY samsung will have some competition again and prices will come down. So much for that dream
Optane is disappointing (Score:2)
The hype gave way to reality and the devil you don't know with underwhelming TBW v. flash when price factored in. I've got to hand it to Intel for at least trying something other than flash but Xpoint is fundamentally a lost cause in terms of mass appeal.
I pray for silicon gods to mass-produce super high density MRAM and put flash out of its misery. While the IOPs are impressive using this capacity in any kind of sustained way is basically impossible with current SSD systems. You'd fry the damn thing in