Intel's 3rd Generation X25-M SSD Specs Revealed
by Anand Lal Shimpi on October 5, 2010 12:04 PM ESTWhat's this? The long awaited specs for Intel's third generation SSD? Indeed.
Internally it’s called the Postville Refresh (the X25-M G2 carried the Postville codename), but externally it carries the same X25-M brand we’ve seen since 2008. The new drive uses 25nm IMFT Flash, which means we should get roughly twice the capacity at the same price. While Intel is sampling 25nm MLC NAND today it's unclear whether or not we'll see drives available this year. I've heard that there's still a lot of tuning that needs to be done on the 25nm process before we get to production quality NAND. The third generation drives will be available somewhere in the Q4 2010 - Q1 2011 timeframe in capacities ranging from 40GB (X25-V) all the way up to 600GB.
Despite the Q1 release of Intel’s 6-series chipsets, Intel is listing the new X25-M as being 3Gbps SATA only. The SATA implementation has been updated to support ATA8-ACS so it’s possible we may see official 6Gbps support once Intel has a chipset with native support.
The new drive’s performance specs are much improved. The comparison between old and new is below:
| Intel Consumer SSD Comparison | ||||
| Intel X25-M G2 (34nm) | Intel X25-M G3 (25nm) | |||
| Codename | Postville | Postville Refresh | ||
| Capacities | 80/160GB | 80/160/300/600GB | ||
| NAND | IMFT 34nm MLC | IMFT 25nm MLC | ||
| Sequential Performance Read/Write | Up to 250/100 MB/s | Up to 250/170 MB/s | ||
| Random 4KB Performance Read/Write | Up to 35K/8.6K IOPS | Up to 50K/40K IOPS | ||
| Max Power Consumption Active/Idle | 3.0/0.06W | 6.0/0.075W | ||
| Total 4KB Random Writes (Drive Lifespan) | 7.5TB - 15TB | 30TB - 60TB | ||
| Power Safe Write Cache | No | Yes | ||
| Form Factors | 1.8" & 2.5" | 1.8" & 2.5" | ||
| Security | ATA Password | ATA Password + AES-128 | ||
If these numbers are accurate, the new Intel drive should be roughly equal to Crucial’s RealSSD C300 and SandForce SF-1200 based drives. There are many different ways to measure this data however so the numbers may be higher or lower in our tests. Note that performance could also go up by the time drives are available as there's still a lot of tuning going on right now. I'd say that at these performance levels Intel had better be very aggressive with pricing because I'm expecting much better from the next-generation SandForce drives.
Write amplification appears to be more under control with the third gen X25-M. Intel upgraded the total 4KB random writes spec from 7.5TB - 15TB on the G2 to a much higher (and wider) range of 30TB to 60TB depending on drive and spare area.
Intel hasn’t disclosed any information about spare area, but given the huge increase in longevity of the drives I suspect that spare area has gone up as well (at least on the larger drives).
The G1 and G2 drives didn’t store any user data in the off-controller DRAM, the third gen drive changes that. A large part of why the C300 is so quick has to do with its large external DRAM, something Intel has avoided implementing in the past due to the associated risk of data loss. Intel refers to the 3rd gen X25-M has having a power safe write cache, which sounds to me like it has an external DRAM paired with a big enough capacitor to flush the cache in the case of sudden power loss.
Full disk encryption is the next big feature on the Postville Refresh. You get AES-128 support on the consumer drives. I’m guessing there’s a new version of the SSD Toolbox in the works as Intel is also promising Windows based firmware updates.
The new X25-M will be available in both 1.8” and 2.5” versions. The 1.8” drive tops out at 300GB, you’ll need the 2.5” form factor for 600GB.
In addition to the new X25-M there’s a new X25-E due out in Q1 2011. Codenamed Lyndonville, this will be the first Intel Enterprise SSD to use MLC flash. It’s not quite the same MLC used on the consumer drives but rather a modification of the 25nm process that trades data retention for longevity.
Standard MLC will last for 12 months after all erase/program cycles have been consumed. Enterprise grade MLC will last only 3 months after exhausting all erase/program cycles but will instead support many more cycles per cell.
The X25-E improves specs compared to its predecessor:
| Intel Enterprise SSD Comparison | ||||
| Intel X25-E (50nm) | Intel X25-E (25nm) | |||
| Codename | Ephraim | Lyndonville | ||
| Capacities | 32/64GB | 100/200/400GB | ||
| NAND | IMFT 50nm SLC | IMFT 25nm eMLC | ||
| Sequential Performance Read/Write | Up to 250/170 MB/s | Up to 250/200 MB/s | ||
| Random 4KB Performance Read/Write | Up to 35K/3.3K IOPS | Up to 50K/5K IOPS | ||
| Max Power Consumption Active/Idle | 3.0/0.06W | 5.0/0.095W | ||
| Total 4KB Random Writes (Drive Lifespan) |
32GB: 1PB 64GB: 2PB |
100GB: 900TB - 1PB 200GB: 1PB - 2PB 400GB: 1.4PB |
||
| Power Safe Write Cache | No | Yes | ||
| Form Factors | 2.5" | 2.5" | ||
| Security | ATA Password | ATA Password + AES-128 | ||
Larger capacities, higher performance, AES-128 support and comparable lifespans to the old X25-Es are all in store early next year. Note that Intel tests 4KB random write performance differently on enterprise vs. consumer drivers so you can’t directly compare the numbers between the X25-M and X25-E. The X25-E will be 2.5” only.
Intel isn’t the only one working on a controller update. SandForce and Indilinx are both heading towards production versions of their next-generation controllers. I expect we’ll see preview class hardware before the end of the year, with mainstream availability in Q1 2011.
Facebook
Twitter
RSS
97 Comments
View All Comments
iwodo - Tuesday, October 5, 2010 - link
Faster Sequential WriteFaster Sequential Read
Faster Random Read
Faster Random Write
Faster then What? IT IS AT BEST SIMILAR TO Sandforce!
It doesn't matter if it is faster then Gen 2, The current King of Performance is Sandforce!
Yeah Right.
Chloiber - Wednesday, October 6, 2010 - link
Faster than Sandforce - in EVERY discipline. I'm talking about real world performance, not completely unrealistic ATTO bench results.PC911mickster - Tuesday, October 5, 2010 - link
Great read Anand!Thanks!
No1uNo - Tuesday, October 5, 2010 - link
The numbers provided are quite intriguing and suggest that the -M drive might actually be a better enterprise drive in many situations. You'd mentioned that "Intel tests 4KB random write performance differently on enterprise vs. consumer drivers so you can’t directly compare the numbers between the X25-M and X25-E." This may be true, but presumably you can compare across generations:The X25-M and X25-E have similar performance in tests presented here - 4k random read used disk = 35.8 vs. 35.4 MB/s for the X25-M and X25-E respectively. The G3 drive is quoted as offering almost 5x that of the X25-M whereas the X25-E performance only improves by about 50%.
Taken together with the identical sequential and random read speeds and the nearly identical sequential writes (170 vs 200 MB/s), database operations and web servers would seem much better served by the X25-M. Somehow I find that conclusion unbelievable. I just can't believe that the difference in random writes would be enough to tip the balance. I must be missing something.
On the subject of specs, I notice that the Corsair Force is faster or identical in all benchmarks vs. the X25-E but on the AnandTech Storage Bench, the X25-E wins. There is clearly more to the benchmarking story.
Mathieu Bourgie - Tuesday, October 5, 2010 - link
A 0.075W power consumption at idle is fantastic, but the 6W at load is a deception, although it makes sense, since it's memory and the more of it there's it, the more power it requires.Seeing as density is doubled from the second generation to the third generation, it makes sense that the power consumption at load doubles. Then again, most of the time, in my case anyway, the drive will be idle, so this might not be such a bad thing after all.
I'm considering a SSD for my next laptop (Buying sometime next year probably, waiting on AMD/Intel new platforms) and battery life is very important to me, so I'd love to see some battery life testing on a laptop, comparing a 5,400/7,200 rpm HDD, G2 and G3 SSD if possible, when those SSDs come out.
chrysrobyn - Wednesday, October 6, 2010 - link
Why would doubling the non-volatile memory necessarily involve a significant power increase? We're not talking about DRAM which needs to get refreshed every few hundred microseconds, nor are we talking about SRAM which has 6 transistors leaking 100% of the time, in use or not. This is all Flash, which will retain state even if you turn the devices off. Doubling the capacity involves increasing the amount of peripheral logic and drivers and such, but that is by no means the dominant factor in active power. Shrinking lithography will decrease the length of wires between features, and smaller devices themselves have less capacitive load, so maintaining one size should even decrease power even acknowledging that smaller transistors may have more off current.Makaveli - Tuesday, October 5, 2010 - link
I was expecting this drive to be 6 Gbps and just run at 3 Gbps until intel released their chipset. So this means we will see the refresh of these drives supporting it, not a huge deal but unexpected. I also would have thought write speeds would have been slightly higher tho still not that important for an OS drive.So it looks like pricing and capacity will be there big advantage with this gen.
Being a current 160GB G2 owner I really don't see a need to upgrade to one unless I have some give me a good price for my used drive.
My usage pattern is also not the same as Jack FX. I've owned my drive since dec 08 and only have like 820GB's of host writes so I think this drive will easily outlast any hard drive that i've owned.
So pro's
Great pricing
Capacity
increased durability
Reliability
AES
Cons
No 6 Gbps Sata
Writes speeds could have been faster
bigboxes - Tuesday, October 5, 2010 - link
As SSDs increase in storage capabilities they further encroach on traditional HDDs place in IT. Anand, how much longer until SSDs are affordable for file storage and practicality (>1TB)? Also, what is the lifespan of data on an SSD? Is there any degredation or data loss over time?Makaveli - Tuesday, October 5, 2010 - link
Owned the drive since December 2009linuxpro7477 - Tuesday, October 5, 2010 - link
I've owned both the Intel X-25M and the SandForce-1222 (Patriot Inferno and Corsair Force series). My experience is the SandForce has some VERY serious bugs, and even if it somehow works properly for you, it's still noticeably slower than the Intel X-25M in actual desktop usage. Sequential transfer speeds are only meaningful when you're transferring huge files. For 98% of your average computer usage for an average user, small (4k) random read/writes are going to have a MUCH bigger impact on "speed" that you can see.And the bugs I mentioned are very big ones. When I put the SandForce drives in my computer, I was getting BSODs and random crashes constantly. That was on a brand-new Intel server mobo with dual Core i7's, not some cheap MSI board. SandForce doesn't want to admit that there are some major problems with their firmware, and as a result, the vast majority of people buying those drives are now having to deal with problems that make the drive unusable and being given no more info than "a new firmware should be available soon." Pfffft.... for $350, I expect better than that. Haven't had a single problem with my Intel X-25M, not to mention the improvement in performance over the SF-1222 drives. And the cost is much better as well. Again, the Intel X-25M G2 has been, for me at least, SIGNIFICANTLY and noticeably faster than the SandForce drives in every scenario other than copying a HUGE file (such as a DVD iso) to another SSD.
Don't buy into the marketing hype behind the sequential transfer rates that are advertised for the SandForce drives. Do some actual benchmarking on your own and you'll find that the SandForce only reaches those benchmarks in very specific situations which you will rarely, if ever, see on your actual desktop. SandForce's claims about higher speed are a joke to the point of being borderline fraudulent, and the online reviews regarding those drives have opened my eyes as to which reviewers out there can be trusted, and who is just copying and pasting marketing press releases and calling it a review.
Anand earned my respect with some of his coverage on the SandForce drives. This was one of the few sites to point out some of the MAJOR problems with the SandForce controller and to actually look beyond the empty promises of 285MB/s sequential transfer rates. People need to understand the difference between large sequential I/O and small random I/O, and how that impacts desktop performance. If more people knew that less than 2% of your average desktop use will involve large sequential transfers, it might help consumers make more educated purchases.
Also, just picked up some Intel X-25E 64GB drives for some servers at work, and the results have been incredible so far. SLC Nand, FTW!! Performance is on-par with ramdisk for every scenario I've tested: high-load DB handling 3,000 transactions per second, a Zenoss Core install monitoring 5,000 devices (so again, TONS of tiny reads/writes), and our corporate email gateway which is doing a ton of filtering (spamassassin, pyzor, razor, clamav, etc) and handling about 500-1000 messages per second. Those are just a few scenarios where using SSD has given us dramatic performance improvements over HDD, by a signficant order of magnitude. We finally have non-volatile ramdisk!!