Intel Responds To X25-M Fragmentation Issue

Vigile writes "In mid-February, news broke about a potential issue with Intel's X25-M mainstream solid state drives involving fragmentation and performance slow-downs. At that time, after having the news picked up by everyone from CNet to the Wall Street Journal, Intel stated that it had not seen any of these issues but was working with the source to replicate the problem and find a fix if at all possible. Today Intel has essentially admitted to the problem by releasing a new firmware for the X25-M line that not only fixes the flaws found in the drive initially, but also increases write performance across the board."

Comments

Least we could do for the readers!

[AllynM]

Guys,

You're welcome :).

Kidding aside, it was great to have a manufacturer as large as Intel work with us and have something good come from it.

Allyn Malventano
Storage Editor, PC Perspective

Re:Good for them

[drinkypoo]

Makes an interesting contrast to intel's response to the FDIV bug, eh? Between this and the whole linux driver thing I'm almost inclined to suspect that intel has learned that you have to serve your customers.

Re:Why'd they call it that?

[QuantumG]

The SWIFT and other banking networks still use x.25. It's a rule of information technology that nothing is ever thrown away.

Re:Good for them

[elashish14]

Agreed. Owning up to your mistakes, whether you're a company or an individual, is a sign of dependability and reliability. I don't know about you, but for me that's a major factor when I purchase something.

Anandtech

[MSG]

On this subject: I finally got around to reading Anandtech's very long article about the current crop of SSD drives. I feel like it was pretty educational, which is good because it took a long time to digest.

In its discussion of performance degradation as drives are used, the article explains that individual pages of NAND memory can't be rewritten. Early in a drive's life, page are remapped when they are rewritten by the OS. As the drive is used, the drive runs out of pages to remap and is forced to copy a block (typically a 512KiB collection of 4KiB pages) to cache, erase the block and then rewrite the block with the new pages. That explains pretty well why write performance degrades, since writing to a block that has data must perform a read and erase operation in addition to the write. However, that explanation also leaves open the question of how the drive prevents data loss if it loses power. Worst case, the OS issues a write and the drive copies a 512KiB block to cache and erases the block, and then loses power. Due to remapping, literally anything could be in that half a MiB. The data loss could corrupt the file that was being modified, obviously, but also any other file on the drive, or parts of the filesystem itself.

I figure there's got to be protection against data loss built-in, but I'm not able to find details regarding any individual drive or manufacturer's approach to solving that problem. Does anyone know more about this subject?

Every SSD has this problem

[Colonel Korn]

We've all read this by now, right?

http://www.anandtech.com/storage/showdoc.aspx?i=3531

The X25 has the same problem as all the other flash drives due to the need to erase in big chunks. Post-slowdown, the X25 is still faster than almost any other SSD that's brand new, and given the same usage, the X25 maintains the huge performance advantage it has from the start. I doubt Intel can really do much to improve this behavior without using TRIM.

I assume their "fix" will be slight tweaking of writing patterns done mostly to fool the mainstream press that had already been acting foolish by picked up this story without noticing the subtleties (such as the problem being present in all SSDs)

Re:Good for them

[adamkennedy]

There's a big difference between admitting to a bug that you can fix with a low/no-cost firmware upgrade, and admitting to a bug which requires a massive recall, and announcing to the market you'll be taking a multi-million dollar loss.

You should be looking at random I/O speeds

[Chris Daniel]

The new OCZ & Samsung drives are faster (and larger) than the X25-M.

For sequential read/write -- yes, they are faster than Intel's offerings. Random read and write operations, on the other hand, are another story. That's one of the biggest issues that SSDs solve versus spinning platters, and no one has gotten it right so far, except Intel.

Re:You should be looking at random I/O speeds

[LordKronos]

I think you misunderstood. Your experience vs. mechanical drives has nothing to do with the issue. The AC said that the OCZ was faster than the Intel. Chris Daniel was simply saying "yes the OCZ is a bit faster for sequential acces, but random access (which is what most users experience the most) is better on the Intel"

If you look at Anand's article You will see that the OCZ beats the Intel slightly at sequential read (about 5%), and by a decent margin on sequential write (slightly less than 3x). However, these aren't things most users typically do...especially the writes. You are only likely to be doing that if you are working with editing large a/v files or something, and since large a/v files take up tons of space, a SSD probably isn't the best candidate for that anyway, given its current cost/storage metric. The OCZ might make sense working in a something like a professional AV editing environment, where you can copy the file off the server, work on it locally, and then copy it back to the server when done.

On the other hand, random reads and writes are something that virtually 100% of users experience on a regular basis, and this is where Intel really shines. On reads, Intel wins by a decent margin (slightly less than 2x the speed, and nearly half the latency). But then look at sequential writes, and Intel really takes the decisive win in that category. While the OCZ is a healthy 4x faster than Velociraptor, the Intel is just shy of 10x the performance of the OCZ (and thus nearly 40x the Velociraptor).

So, when you compare the Intel and the OCZ, the Intel loses slightly and decently on 2 operations that are less common, and it wins decently and decisively on 2 operations that are more common. Thus it's a pretty good stretch to try and say the OCZ is faster than the Intel.

Customer Experience

[robvangelder]

I once dined at a restaurant that took my order, but minutes later realised they couldnt make it due to stock shortage. I got a different meal, and they told me mine was for free!

The way a company recovers from a problem can actually turn into a net positive experience for the customer.

In my case, I'm turned from an unsatisfied customer, to an advocate. For sure, I've recommended friends dine there since then.

Every interaction is an opportunity to delight the customer. Even those events that at first feel like a disaster unrolling.

Allyn Malventano Theory - It Can Write Faster!

[JakFrost]

The most interesting thing is the last section on the last page.

PC Perspective: Intel Responds to Fragmentation with New X25-M Firmware - My Theory - It Can Write Faster
by Allyn Malventano 2009-04-13

My own personal theory is that Intel got things *too right* with their custom controller. ...

Despite using MLC flash memory, competitors have broken the 200 MB/sec sequential write speed barrier, and have done so with only 4 channel controllers. The X25-M talks to its flash across 10 parallel channels. If the X25-M was truly flash speed limited at 80 MB/sec, other MLC flash would have to be over 6x as fast to achieve stated speeds over the fewer channels available. ...

My hunch is they expected MLC write speeds to remain relatively low across the marketplace, and like many other products in similar chains, imposed a hard limit of 80 MB/sec to their M series drives. ...

If an M series drive could write as fast as an E series drive, there would be considerably less market for the latter. ...

I just think it can go faster than 80 MB/sec.

I think that Allyn is onto something because if you look at the graph for write speed of the X25-M (MLC) it seems utterly perfect at 80 MB/s, almost like there is an artificial cap on the speed, while the one from the X25-E (SLC) series it produces a standard waveform, like Allyn pointed out, and not an artificial flat line.

I too believe that Intel is artificially capping the performance of this drive and they might decide to uncap it sometime in the future once the competitors start snapping at their heels or if enough time goes by and they decide to introduce a new SSD MLC based performance/server oriented product line and remove the cap then. This is very similar to the situation with processor multiplier locks that they remove in their performance oriented Extreme processor lines.

I frankly don't like this kind of behavior from Intel since they know that they have the upper hand so they are just doling out enough performance to beat the competitors and to satisfy the current customers but at the same time holding back to create a market for their X25-E product line with slightly higher performance.

I think the other shoe will drop sooner or later on the 80 MB/s cap.

Research

I've been doing research into Solid State Disks in the last few weeks and this article is yet another one of those for Required Reading in the course of learning about SSD. I've even wrote a detailed post with links to reviews and articles. You can read up on the linked articles to get a good primer on things.

Solid State Disk Benchmarks