Overclock.net banner

[TH]Intel 3D XPoint Optane SSD DC P4800X Tested

3K views 22 replies 18 participants last post by  guttheslayer 
#1 ·
Well, they finally got their hands on Optane! Looks impressive
smile.gif
Certainly the fastest, ever!
Quote:
The elusive 3D XPoint non-volatile memory technology is finally ready. Well, kind of. You see, we don't actually have it set up in our lab yet. Intel has a keen interest in keeping its new progeny under the strictest of wraps, so we weren't actually allowed to take possession of the 3D XPoint-powered 'Cold Stream' Optane SSD DC P4800X. Instead, the company offered us an opportunity to remotely test the drive in a top-secret lab tucked away on its Folsom, CA campus. Of course, we had to get our hands on the hottest storage device ever, if only virtually. And in the end, we came away with some impressive preliminary numbers.
http://www.tomshardware.com/reviews/intel-optane-3d-xpoint-p4800x,5030.html

 
See less See more
2
#3 ·
wonder if there are any prosumers who will opt for this over a 960 Pro 2TB

What are the chances X299 and X399 platforms will support Optane? Would be nice to get in on the hype once they are more mature without having to build a new system.
 
#6 ·
Quote:
Originally Posted by CrazyElf View Post

That said, it is a very, very long way from taking on RAM. It's better than NAND, but not nearly as good as DRAM. I'd love to see them include a RAMDisk for a comparison test.
My RAMdisk has about 2x 4k random performance in QD1, but half the 4k random performance in high queue depth. Desktops dont really see much queue depth though so that difference will only matter in server situations. In sequential testing my RAMdisk is 4-5x faster in low and high queue depth situations than the Optane drive.
 
#7 ·
Quote:
Originally Posted by twitchyzero View Post

wonder if there are any prosumers who will opt for this over a 960 Pro 2TB

What are the chances X299 and X399 platforms will support Optane? Would be nice to get in on the hype once they are more mature without having to build a new system.
375GB at what was it $1500? Or 2TB on the 960 Pro for less...

1000 times this 1000 times that so far it doesn't seem that much faster than other SSDs, the only thing 1000 times so far seems the price.
 
#10 ·
Quote:
Originally Posted by nakano2k1 View Post

SSDs are already more than what 95% of users need to begin with. This whole rush for speed is nothing more than corporate cash grab. "OMG you need 1.5gb read speeds"
rolleyes.gif
These are not targeted at average joe consumers...
 
  • Rep+
Reactions: ropesend
#11 ·
This is Intel's first gen remember first gen data drives lol give them the room to make xpoint what it shoulda been unless all that marketing bs was just to get hype rolling then fail Intel. But I really hope Zen+ gets optane support from micron and they make the dimms that'd be a nice upgrade. Zen+ 16-32 cores micron quantex dimms put in like 256 GB worth toss in a nice huge raid 5 SSD array and I'd be set till next gen of goodies come out.
 
#12 ·
Many on OCN say INTEL does not innovate anymore. This is one example on how they are pushing technology further. Do people recall that originally the fastest SSDs came from INTEL and not the industry giants? They pushed SSD controllers to a new level and now we got products from Samsung and Co which we cannot live without.

Intel has the resources to make this technology as standard and , comparing to AMD, they will manage to push it to the average consumer in the coming years by making a profit. They have to recover their investment and we will pay a lot.

This drive is also a warning that the race to increase IPC for performance is not what it was. There are various other components where improvements have been quite small in the last decade.
 
#13 ·
Agreed with Wish.
Quote:
Originally Posted by Wishmaker View Post

Many on OCN say INTEL does not innovate anymore. This is one example on how they are pushing technology further. Do people recall that originally the fastest SSDs came from INTEL and not the industry giants? They pushed SSD controllers to a new level and now we got products from Samsung and Co which we cannot live without.

Intel has the resources to make this technology as standard and , comparing to AMD, they will manage to push it to the average consumer in the coming years by making a profit. They have to recover their investment and we will pay a lot.

This drive is also a warning that the race to increase IPC for performance is not what it was. There are various other components where improvements have been quite small in the last decade.
Agree with Wish. There are certainly many tangible benefits I'm seeing with my Intel 750 series PCIe x4 drive compared to a regular SSD. This only goes even faster from the looks of it.
 
#14 ·
A lot of you guys seem to be looking at this wrong. These don't need to be faster than DRAM, they just need to be fast enough, and more importantly high capacity and cheap enough, to change the way we use DRAM.

The traditional paradigm was to read all data that must be used off of the disk and copy it into DRAM. It is then worked with in DRAM and then kicked back out to make room for the next data that is being read off the disk. If the previous data needs to be used again, it has to be loaded back off the disk again and into DRAM. This system works, but isn't terribly efficient.

In the last few years, we have finally gotten enough DRAM that it is feasible to preload the majority of data that might be used for a particular program into DRAM and leave it there until that space is needed for something else or the program is exited.

This helps considerably because we aren't wasting as much time reloading the same data over and over, but even if we got to the point where we could have enough DRAM in a system to load the entire disk into DRAM upon booting the system (which, frankly, we will probably never get to), we would still hit that major bottleneck of disk read speed. Reading the entire disk into DRAM would take quite a long time, several minutes or even hours depending on the disk size and speed, every single time the system was powered back on. It would also leave all your data in a vulnerable state unless all changes were mirrored back onto the disk, which would be an additional bottleneck.

Around the same time, we got SSDs, which also help considerably because they cut down on that major bottleneck. They can serve data into DRAM much quicker than HDDs, but unfortunately, they are still nowhere fast enough to actually remove the bottleneck. They do however make pre-caching (relatively speaking, obviously caching the entire disk would still be absurd) large amounts of data more feasible. It is a step in the right direction, but that bottleneck is still there forcing us to flip flop data between disk and DRAM constantly, which in turn forces us into needing larger and larger capacities of DRAM for all that caching.

If Optane can provide us with a disk that is large and cheap enough to replace SSDs and HDDs, while living up to its claim of being within an order of magnitude the speed of DRAM, it could flip that whole scheme on its head.

SSDs with their hundred thousand times higher latency than DRAM aren't going to work for all but the least demanding of data processing workloads (streaming audio or video up to a certain resolution, and working with text, is probably about it).

HDDs are even worse at four million times higher latency. They're basically completely useless without DRAM.

Optane at only 10 times the latency? Most daily tasks could be read and written directly from/to the disk, skipping DRAM entirely. Without having to expand DRAM capacities at all, we would suddenly have 3-4 times the effective DRAM capacity, because all that space currently taken up by low-priority latency-insensitive data would be freed up.

This would compound with the fact that when data is transferred between DRAM and disk, the bottleneck will be lessened by many orders of magnitude, just as it was when switching from HDDs to SSDs, only even greater.

Furthermore, because less data needs to actually be cached into DRAM, the frequency at which that bottleneck will occur will also significantly decrease.

So you see, Optane doesn't need to replace DRAM to be a game changer, it just needs to be fast enough to start synergizing with DRAM instead of holding it back like current disk technology does.
 
  • Rep+
Reactions: The-Real-Link
#16 ·
XPoint SSDs sure look beastly.
 
#18 ·
The day Optane replaces DIMM for both Storage drive and RAM is the day everything can be finally integrated (Meaning less component to assemble a PC)

It means console will get better in closing the gap between them and the high end PCs.

I dunno, maybe CNT and Optical Graphene is the final keys to make that happens?
 
#19 ·
There's that study they finally made graphene in a explosive way that made it possible to mass produce. But I thought they were leaning more toward optics using light to transmit data. But whatever it is they'd better start sampling soon if they wanna keep making smaller and smaller nodes. Soon be submitted atom sized lol.

That pic is the fpga board for optane when they finally got optane outa vapor ware territory and had a actual product.
 
#21 ·
Quote:
Originally Posted by CrazyElf View Post

Check out that random read at QD1. Latency seems to be pretty good too compared to NAND. Not nearly as good though as Intel claims though:
I don't think Intel is actually wrong, just a bit misleading. I bet the latency will be nanosecond-scale once we get new CPUs with IMCs that natively support them and Optane-based DIMMs. Right now, it's going through an NVMe controller, but direct access via the CPU should help.

Quote:
Originally Posted by guttheslayer View Post

Now before going so far off...

When is the consumer optane coming?
http://www.anandtech.com/show/11227/intel-launches-optane-memory-m2-cache-ssds-for-client-market

wink.gif
 
#22 ·
Quote:
Originally Posted by guttheslayer View Post

Now before going so far off...

When is the consumer optane coming?
Allyn at Pcper said 'towards the end of the year' which can actually mean 'early next year to be able to buy one'.

Looking at the way the enterprise version of 750 was priced, it was what, x3 the price of 750? Take $1500 and divide by 3 we get $500. Expensive, but not insane considering the low qd random read perf.
 
#23 ·
Quote:
Originally Posted by Darkwizzie View Post

Allyn at Pcper said 'towards the end of the year' which can actually mean 'early next year to be able to buy one'.

Looking at the way the enterprise version of 750 was priced, it was what, x3 the price of 750? Take $1500 and divide by 3 we get $500. Expensive, but not insane considering the low qd random read perf.
That will be $100 for 75GB version or close to $180 for 128GB version.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top