Intel Explains 20nm NAND Endurance Concerns on the SSD 335
by Anand Lal Shimpi on November 17, 2012 1:49 PM EST- Posted in
- Storage
- SSDs
- Intel
- Intel SSD 335
It's common knowledge by now that as NAND cells shrink in size (thanks to smaller process nodes), their endurance and program/erase latencies both suffer. Consumers wouldn't really be happy with newer drives dying sooner and performing worse than their predecessors, so controller and NAND makers have to work extra hard to compensate for losses due to the physics of NAND.
In the second half of this year we saw the beginning of a transition from Intel's 25nm MLC NAND to a 20nm process. The smaller process will eventually allow us to have larger SSDs with 16GB NAND die (up from the current 8GB max), but it should also help drive SSD prices down as you can fit more 20nm NAND cells per 300mm wafer than you can at 25nm. The cost motivation alone will move production from 25nm to 20nm in fairly short order. The question remains: how is endurance impacted by the move to 20nm? When Kristian reviewed Intel's SSD 335, he tried to find out.
The 335 is Intel's first branded drive to ship with 20nm 2-bit-per-cell MLC NAND. Like most modern Intel SSDs, the drive reports total NAND writes as well as the percent remaining program/erase cycles on the NAND. By looking at both values you can get a general idea of the expect lifespan of the NAND in number of program/erase cycles.
In his 335 review, Kristian calculated the endurance of Intel's 20nm NAND on the 335 to be below 1,000 program/erase cycles. Intel confirmed that its 20nm MLC NAND is rated at 3,000 p/e cycles and that Kristian's results shouldn't have happened. Others duplicated the results around the web, so we waited for an explanation from Intel. Today we have that explanation.
The Media Wearout Indicator (MWI) on Intel SSDs looks at total number of times the drive's NAND has been cycled and then divides it by the p/e rating for the NAND to determine its value. The p/e rating is hard-coded into the firmware for the NAND. The 335 we reviewed had the p/e rating set to 1500 cycles, which was accurate for an earlier, non-production version of Intel's 20nm NAND. The 335 should have shipped with firmware that set this value to 3000 cycles, but someone forgot to set the variable to the right value in the firmware code. Woops. As a result, the MWI value on the 335 decreases at 2x the rate it should. This doesn't mean the drive is wearing out twice as fast as it should, just that the MWI data is inaccurate. Take our numbers from the 335 review and double them to get how long the 335 may last.
MWI from our 335 review, incorrectly low
Kristian's review showed that to be roughly 250TB of writes, which means the actual value is aroung 500TB of actual NAND writes (incompressible). Doing the math on the 240GB capacity gives us 2083 full drive writes over the life of the drive, or about 5.7 years of useful life if you write 240GB of data to the NAND every day. Even if your workload has a write amplification factor of 10x, you're still talking about 24GB of writes per day for nearly 6 years.
The math works out to be around 1500 p/e cycles, however remember that when the MWI hits 0 the NAND isn't truly exhausted. It should last well beyond that point. The MWI hitting 0 on an Intel drive is just a good point to begin looking at replacing the drive.
Intel will have a firmware update for the 335 out before the end of the month that fixes the MWI reporting behavior, but if you're concerned about endurance on the 335 - I wouldn't be.
There's another issue however. The 335 should have similar endurance to the 330, but even with our revised numbers the two look very different. It turns out MWI reporting on the 330 is not working at all; the MWI value will never drop, regardless of how much you write to the drive. Intel has committed to fixing this issue in a December firmware update for the 330.
Neither issue fundamentally impacts the functionality of the drive or its endurance. But if you're closely monitoring the MWI values, you should keep all of this in mind.
That's all for the Intel SSD update. I've been hearing more reports of dying Samsung SSD 840 Pros and I believe I know the cause (firmware related, should be fixed in the latest shipping revision) but I'm still waiting for confirmation on one last thing before explaining what's going on there.
Facebook
Twitter
RSS
31 Comments
View All Comments
mmonnin03 - Tuesday, November 20, 2012 - link
As the process nodes shrink there are fewer and fewer cells to hold the charge in a NAND cell. And as you shrink there is a smaller gate oxide (area) and often times thinner oxide trapping that charge in the cell. Also keep in mind for MLC, it must differentiate between 4 possible voltages in the cell. 0 and 1 for 2 bits. For TLC it goes up agains. So as process nodes shrink it becomes harder to hold the charge and harder to read that charge, thus it becomes slower as ECC grows exponentially.At some point I don't see NAND being what we want it to be in terms of reliability and cost. The number of Immersion scanners need to produce it will raise chip costs and the reliability will be to low.
Beenthere - Saturday, November 17, 2012 - link
Rushed to market SSDs - for profit over performance and reliability. Who knew? Answer: Anyone with a clue.Not to worry, five or ten firmware updates should make the latest SSDs almost usable. Good Lord there must be two suckers born every second.
Oh yeah, don't forget that Anand himself said: "SSD technology is immature..." and that "...consumers should wait 6-12 months before buying". That fact applies to the current trick-of-the-week, rushed to market SSDs from Intel and Samsung.
If people refused to buy this crap then the SSD makers will spend more time validating the hardware or lose sales.
BTW, as we see from the story above Intel's SSDs are no better than Samsung's or many of the other brands - most of which have issues of one sort or another - because they have not been properly validated.
Meaker10 - Saturday, November 17, 2012 - link
Intel has left the toilet seat up where as others have left the front door wide open when leaving.This is a basic counter with no bearing on functionality so I have no idea wtf you are going on about.
MrSpadge - Sunday, November 18, 2012 - link
+1seapeople - Sunday, November 18, 2012 - link
What he's going on about is that companies should release these drives with "BETA TEST VERSION, USE AT OWN RISK" tattooed all over the box for at least the first 3 months, maybe 6 months for the worst brands.boshi - Sunday, November 18, 2012 - link
I don't know if this is enough to make the assertion that "Intel's SSDs are no better than Samsung's or many of the other brands".This is an issue that many ( who would not even know that the MWI existed ) would never encounter, and it happened on a review sample running a pre-release firmware.
I'm not sure that anyone reporting on SSD failures have a large enough sample size to report anything useful about their failure rates.
Beenthere - Sunday, November 18, 2012 - link
Intel has not demonstrated that their SSDs are more reliable or compatible than any other brand of SSDs. They have had issues with their SSDs as have most other SSD makers - because the drives are never properly validated but instead are rushed to market for great profit.The fact that unscrupulous SSD makers continue to rush unvalidated products to maket for naive consumers to buy and become un-paid Beta Testers, is a disgrace. Anand was more tacful, but the fact remains that many consumer grade SSDs are simply not ready for Prime Time.
It's a bad joke to see SSD makers introduce a new series of SSDs every month or two. They haven't even corrected the problems with the last 3 SSD series but they continue to crank more garbage out for gullible consumers to lap up.
jwilliams4200 - Sunday, November 18, 2012 - link
Anand:You really need to add a "report SPAM" button to your comment software.
The SPAM comments seem to stay up for quite a while.
JarredWalton - Monday, November 19, 2012 - link
The admins (e.g. me, Anand, Ryan, and some others) have a Spam button that we can use to delete comments. It might take a day or two before we notice, but we'll get it eventually. If you really want to report spam, though, email me: jarred.walton@anandtech.com. Give me the article name and the user name and I'll usually be around to delete it within the hour (and sometimes within minutes). Thanks!bim27142 - Sunday, November 18, 2012 - link
so MWI for my 330 was inaccurate and i guess this was something that is not to be truly happy afterall...