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

Good for them

[mc1138]

I'd much rather have a company own up to an issue, fix it, and move on, rather than deny it or try to use PR to quiet it away.

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:Good for them

[Bullseye_blam]

Please gents, think logically here. This is something fixed ~three weeks after reproducing the problem. And it was something that could easily be fixed with a firmware update. It's not like the hardware was broken.

So, good job Intel for fixing it, but patting them on the back for admitting a problem [on a tiny user base] that was easily fixed is a delusion.

-bullseye

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.

Re:Good for them

[drinkypoo]

So, good job Intel for fixing it, but patting them on the back for admitting a problem [on a tiny user base] that was easily fixed is a delusion.

You're inferring things from my post that aren't implied. What I'm saying is that intel is perhaps no longer pure, concentrated evil -- not that I want to go start sucking dicks in the executive washroom.

Re:Good for them

[Hurricane78]

How do you know, that is was easy to fix? Did you fix it? Do you know someone who fixes it?
Or are you perhaps talking out of the wrong orifice? ^^

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.

Re:Good for them

[afidel]

The fact that it was fixed AND QA'd in 3 weeks proves it was easy to fix! If you've ever worked for any large company you know three weeks is kind of the minimum for this kind of thing, you need a couple days for meetings to discuss the problem and brainstorm problems, a few days to formulate actual solutions, a few days to test, a few days for QA (minimum) and then a day or two to package it up, get with the outside content providing group and then hand something over to marketing.

Re:Good for them

[SWPadnos]

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 charging them more for '486 chips until you pay for the replacements.

There, fixed that for you.

Re:Good for them

[MichaelSmith]

I am betting they had the new version ready to go and held it back because they were afraid of regression. When it became apparent they had to go forward they decided to take the risk. Thats why it only took three weeks.

Re:Good for them

[nanospook]

Maybe they didnt want to publish this small fix so soon in case a major problem becomes apparent and they actually have to recall the drives physically.. now I'm just being paranoid..

Re:Good for them

[hairyfeet]

Well to be fair to Intel as you can see here the odds of anybody hitting the bug(hell the odds of Intel accidentally hitting the bug) were pretty much slim to none. Nobody(including Intel) would have probably ever found out it even HAD a bug if Thomas Nicely hadn't written a program to hunt for primes and ran it on a Pentium. Let's face it: It was 1994. Most folks were running simple spreadsheets and simple games on Windows 3.11 at the time. The odds that they would have actually been doing enough floating point number crunching to actually hit the thing was about the same as hitting the lotto while being struck by lightning.

So to be fair it wasn't like the early 360 where they knew they had a turkey and pushed it out the door anyway, or Nvidia where they pretended there wasn't anything wrong as they passed out faulty chips that fell apart. They should have handled the PR better but the odds that the average user would actually hit this thing really were pretty damned remote.

Re:Good for them

[eebra82]

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.

That would only work in a perfect world. It's like when IBM admitted to the scratching disk problems it had a few years ago. Even if they admitted the problem fairly early, it didn't stop people from dropping the brand.

In reality, if Intel admitted the problems, it would go from a rumor/forum discussion to public announcement with worldwide dirt on the company's drives. Furthermore, we don't really know how many drives are really affected by this problem. I have two X-25M disks myself and have not encountered any problems at all.

So if we look at it from their side, they may have had a problem which - for all they knew - was only limited to a very low percentage of their total shippings. Why would they then want to go public with it and cause a major upset? From a corporate's point of view, it would make no sense.

Re:Good for them

[prisoner-of-enigma]

The fact that it was fixed AND QA'd in 3 weeks proves it was easy to fix!

You're assuming it's really fixed. Or that it was fixed without introducing new, yet-to-be-discovered problems. Technological history is replete with examples of "quick fixes" that ultimately had ancillary negative impacts. What if Intel fixed the issue by altering the wear leveling algorithm...thus shortening the projected life of the device? We wouldn't know about it until much later, and by then it would be too late.

Note I'm not saying this is the case. Far from it. I'm just saying that it comes down to two possibilities:

- it was a quick fix that will have no detrimental effects to the product, which somehow magically escaped Intel's legendarily effective quality control processes and made it into a flagship product.

- it was not a quick fix, it was a rushed fix that may have not solved the problem (it may only delay the onset) and may ultimately cause other issues we have yet to encounter or even understand.

Re:Good for them

[prisoner-of-enigma]

Let's face it: It was 1994. Most folks were running simple spreadsheets and simple games on Windows 3.11 at the time. The odds that they would have actually been doing enough floating point number crunching to actually hit the thing was about the same as hitting the lotto while being struck by lightning.

While you're correct that "most folks" were not going to encounter the bug, the very people that needed the (then) high-end performance of a Pentium were the ones most likely to encounter it. I was rendering 3D animations on 3D Studio for DOS back then, and it was amazingly heavy on the FPU.

Re:Good for them

[prisoner-of-enigma]

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.

But that's the point: Intel hasn't owned up to any mistake. It's issued a new firmware with the nebulous comment that it "increases performance." There's no mention of it fixing anything that was wrong. Intel remains publicly mute on anything being wrong with the prior firmware despite numerous benchmarks and tests showing otherwise.

If this is "owning up to your mistakes" then I'm going to have to change the definition of the phrase.

Re:Good for them

[drinkypoo]

- it was a quick fix that will have no detrimental effects to the product, which somehow magically escaped Intel's legendarily effective quality control processes and made it into a flagship product.

And I have to say again, you mean like a math bug in their spanking new processor? You're acting like there's no precedent for something like this happening inside of intel, when in fact the opposite is true.

Here's a third possibility: intel knew the problem existed, but shipped the devices anyway because they figured that nobody would hit the problem before they managed to get a patch out the door -- possibly because they had a deliverable involving shipping a certain number of units, to satisfy some customer who may have bankrolled part of the development, or perhaps had just promised to be some insanely large number of units if intel could get them out by a certain date (which is nearly always madness because it so often leads to a scenario like the one I've just described.)

Note that I'm not promoting any particular view, only accusing you of a lack of imagination and a failure to remember history.

Re:Good for them

[hairyfeet]

That is why i thought their second recall idea was the way to go and should have been done first. For those that don't remember they originally wanted folks to PROVE they needed floating point, but when folks had a shit fit(and rightly so) they simply said if you were willing to pack up your CPU and ship it to us we will send you a replacement. The simple fact is most folks didn't bother and the ones that did were guys like you that NEEDED floating point.

So in hindsight I would say the only problem Intel had was the PHB that came up with the first recall idea, as it was a perfect example of what NOT to do. Always make it seem to your customers you are trying to give them a fair deal, even if it benefits you. I once had a teacher that used an example from his own work experience that made a perfect example of this. He was running a large chain of stores in southern Florida(Fred's I do believe) and the corporate policy was to NOT take back certain items that the customer may have worn, like say bikinis. He said "screw the policy, give 'em a gift cert for the price of the item and then chunk the item in the trash."

Why did he do that? Because he was SMART. He saw that those with gift certs always spent MORE than the cert was worth, because they would always end up finding things they wanted while looking for what to spend the cert on. He said that sales for the stores he was running picked up nearly 30% just from the extra cash and goodwill generated by the simple change in policy. The same thing would have applied here if Intel had simply handled it better, as the odds of most folks caring about the FDIV bug enough to pull their CPU and wait for a return would be pretty slim, and by handling it better and adding a little spin they would have generated good press for themselves. Always give your customers the impression that you are trying to give them a fair deal and the goodwill generated will return to you as increased sales. Make the customers feel you are trying to screw them and watch it come back to bite you in the ass.

Re:Good for them

[LordKronos]

they simply said if you were willing to pack up your CPU and ship it to us we will send you a replacement.

most folks caring about the FDIV bug enough to pull their CPU and wait for a return

FYI. I exchanged my P60 when that bug hit. You imply that you had to give up your processor for a few days, but that wasn't the case. They took a credit card number and placed a hold for the price of the processor. They then shipped you the new process, and you shipped back the old processor after swapping it out (and I believe return shipping was prepaid, but I'm not positive). If they didn't receive the old processor they would just make the charge against your credit card.

Re:Good for them

[Anonymous Coward]

Here's a third possibility: intel knew the problem existed, but shipped the devices anyway

This is most likely, and it's really a question of spin as to how big the problem was. PC Perspective (the website which 'discovered' it) tried to play it up a lot; it was their big story and they wanted to ride it. But it was clear from other reviews (and even a careful reading of PCPer's description) that it wasn't likely to strike in real world use. You have to really work at it to get the old firmware into the low performance state. It wasn't the end of the world either: even in the degraded state, the drive's random IO performance was still blowing the competition away. The only thing which I wouldn't have liked living with was the hit to sequential write performance, which was cut in half more or less IIRC.

This was a non-crippling no-data-loss performance bug which would affect relatively few people. If you're surprised that people ship software (firmware, natch) with a known performance bug so benign as that, boy have I got a bridge to sell you. The thing to be afraid of is vendors shipping with data loss bugs...

Re:Good for them

[hairyfeet]

That is fine if you HAVE a cc, a lot of folks I knew at the time didn't. And of course even fewer of those that HAD a cc AND had a Pentium affected by the FDIV bug had the skills required to change out a CPU on their own. So for them it would have been -contact Intel, give Intel cc number, wait on CPU, get CPU, have local shop change out CPU and then try to get the original CPU back to Intel before they got charged. For your average person that was just more PITA than it was worth for a bug that frankly the odds of them hitting were slim to none.

Plus you have to remember PCs changed hands a LOT during that time. I remember in the 5 years between 94-99 I went from a 486 to a P60 to a P100 to a P233 to a P2 450 to a P3 550 to a P3 700. And working PC repair at the time I can tell you I was NOT alone in having gone through that many boxes. It seemed like you barely got the thing unboxed before programs were coming along that needed faster than what you had. That is why I am SO glad PCs have long passed "good enough". I am typing this on a 1.1GHz Celeron I picked up in 2K as a cheap refurb and it is still going strong on Win2K 9 years later. Makes a great Netbox. My P4 3.6GHz will be handed down to my oldest next month when I build me a new duallie. He will probably get ANOTHER 5 years out of it before it gets handed back to me to retire the 1.1GHz Netbox.

But back then things were different. Everyone seemed to be hopping from one PC to another as speed and RAM kept leapfrogging and prices plummeted. So many simply wouldn't have bothered as they had already let the affected Pentium go for a new PC that was in every way better than the one they had. I'm just glad that those days are finally behind us.

Oh BTW, I ran into a girl in school a couple of years back that was STILL running a Pentium affected by the FDIV bug. Poor thing was poor as a snake and wanted to give it to her kid as she used her student loans to get a laptop. The teacher called me to the front of the class and asked me in front of her if I could upgrade it, I just smiled and told her "Just wait 3 minutes" and pulled an old Compaq 750MHz out of my car that I had gotten give to me on an office job. I sat her machine by the trash and said "Here you go, on the house and 10 times faster and better than that." Poor little thing bawled like a baby. She just couldn't believe somebody would give her kid a nice machine like that. So what one of us might consider junk can really make somebody's day.

Re:Good for them

[Anonymous Coward]

I'd much rather have a company own up to an issue, fix it, and move on, rather than deny it or try to use PR to quiet it away.

You mean like OCZ did with the Core's and Apex? They just offered huge rebates to dump the old stock on unsuspecting customers...

In Other News..

[TechyImmigrant]

In other news, Microsoft has responded to reports that Windows Vista is slow, buggy, insecure and horribly bloated by releasing DOS 3.2

Re:In Other News..

[voidphoenix]

C'mon guys, this was actually funny.

Re:In Other News..

[joaommp]

Well I laughed. But it was actually off-topic :P

And I'm probably getting modded redundant.

Re:In Other News..

[TechyImmigrant]

I authored that comment. I considered it on topic because the topic is the response of a huge corporation to reports of dysfunction in its product.

The humor was in comparing Intel's quick, honest and effective response to a product dysfunction to Microsoft's failure over decades to respond to the well documented dysfunction in its OS products.

Re:Woah! If it wasn't fast enuf already

[Anonymous Coward]

How much more do you guys need. Intel is the fastest of the bunch and you're getting free speedups at no cost!

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

SuperTalent (i think) is also bringing out a PCIe based SSD (as the fastest SSDs are reaching SATA II speed limitations).

Re:Why'd they call it that?

[TheGratefulNet]

I also go back to the old x25 datacomm days. so I thought 'huh, what? x25 is coming back??'

what a poor choice of names. clearly, intel has some VERY young people working there that have no idea about past things by the same name.

(admittedly, x25 was not as popular in the US as, say, europe)

Fragmentation issue?

[GuldKalle]

At first I read fragmentation as in "frag grenade".
Guess I've been playing too many violent games. Oh, that reminds me - tax reports are due tomorrow, right?

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:Least we could do for the readers!

[AbRASiON]

Get all over the OCZ please, we need as much info as possible.
We're right at the cusp of SSD's becoming reasonably priced for enthusiasts now (not just ultra-rich enthusiasts) and I for one would like to know about the 120 and 240gb OCZ drives anand has dabbled with.

Also future products might be nice too, I am almost positive OCZ will have learnt a lot in the past month, we'll see some seriously good products come out within 3 to 6 months in the SSD scene, I'm sure of it

Re:Least we could do for the readers!

[moon3]

Great research and review.

I read it right through. Thanks to you, the best SSD drive got even better.

Anyway, I am still hasitating to put something that lasts only 10,000 erase cycles into my system..

Re:Least we could do for the readers!

[AllynM]

I touched on this at the end of this page:

http://www.pcper.com/article.php?aid=691&type=expert&pid=8

OCZ is getting there, but they are trying to keep up with the IOPS of Intel's 10 channel controller with their own 4 channel controller. Something has to give. In this case it is their Vertex fragments fairly quickly and won't come back on its own. It *requires* a TRIM utility to be run on it to restore full write speed.

It's a tradeoff. With the new firmware, the X25 goes *slightly* slower with random writes of random sizes, but this is because it is more aggressively defragmenting itself on the fly. It does this to keep closer to its full write speed. The Vertex does not have the luxury of that overhead with only 4 channels.

Re:Least we could do for the readers!

[AbRASiON]

Based on your examination of the situation along with anandtechs and the fact that both OCZ and Intel seem to be aggressively improving these products, it seems to me it might be silly to even consider the X25-M or the Vertex.

Something tells me the SSD scene is moving so fast that within literally 6 months one of these 2 companies (or a competitor taking note) will have a product superior in size, speed and price to those 2 very very soon.

It's a good time to have a little bit of patience I think.
- Scott

Re:Least we could do for the readers!

[edmudama]

The X25-M datasheet guarantees 20GB/day for 5 years. How many DVDs to you torrent each day?

Re:Least we could do for the readers!

[Anonymous Coward]

At the last point in the article you mention that no other drive has as high an IOPs number as the X25 from Intel. I would like to point you to the STEC ZeusIOPS product. While not in the consumer arena it has exceedingly high IOPS.

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.

really

[Pvt_Ryan]

OMG they are admitting to it..

quick stop press, a company has received a bug report, admitted a bug in their software exists & released fix....

And this was /. worth why?

Re:Attention linux users

[Anonymous Coward]

haha, mine is /dev/hda still so nyah nyah

Please, no more smokescreens.

[Daniel Phillips]

This was forseen: Intel will ultimately be forced to redesign their flash write algorithms

The point of this is, please please please if you are an engineering manager, when you make a collective booboo, no smoke screen please! It is unlikely to go unnoticed, and nothing positive will be achieved for you, your company, your potential customers or your tech audience. Instead, just come clean, admit the problem and get busy on the fix. Down that path lies increased trust, whereas the doublespeak path only erodes credibility. I certainly will be double checking any future claims, because of how this played out.

Anyway, big props to the team for implementing what appears to be a superior solution. Hey, how about just open sourcing that firmware and let everybody help make it even better? Just a thought.

Re:Please, no more smokescreens.

[DigiShaman]

Hey, how about just open sourcing that firmware and let everybody help make it even better? Just a thought.

Not going to happen.

They're in a competitive market of bleeding edge technologies such as SSD storage. They will want every advantage they can get. This will require both hardware and software optimizations make their product stand out among the competition.

I'm sure Intel will open up the open sourced spigot much like ATI has with older products. Just don't count on them from cutting edge products anytime soon.

Re:Please, no more smokescreens.

[LordKronos]

Ummm, what you were discussing in that link has nothing to do with what this firmware is fixing. You were discussing performance decreases over time through ordinary use. This firmware fixes a bug that (so far) has only been able to be replicated under certain benchmark conditions, and has not yet appeared under real world conditions.

Don't pat yourself on the back too hard.

Re:Please, no more smokescreens.

[Anonymous Coward]

Hey, how about just open sourcing that firmware and let everybody help make it even better? Just a thought.

Oh yeah let's just give away our secrets to OCZ and other SSD manufacturers for free! Good idea!

Re:Please, no more smokescreens.

[pyite]

The problems which that link discusses are general problems, not Intel's. Even in the worst case, the Intel drive is still better than all the other MLC drives. Anand did a very thorough analysis here and it's probably one of the best mainstream pieces of technical writing I've ever seen.

He basically justifies the whole existence of Anandtech with that one article.

Re:Please, no more smokescreens.

[Daniel Phillips]

Ummm, what you were discussing in that link has nothing to do with what this firmware is fixing. You were discussing performance decreases over time through ordinary use. This firmware fixes a bug that (so far) has only been able to be replicated under certain benchmark conditions, and has not yet appeared under real world conditions.

I have no idea what you are talking about. The issue discussed in the post and the issue addressed by intel in the new firmware are the same by all appearances.

Re:Why'd they call it that?

[Fortunato_NC]

No, that's the exception that proves the rule.

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?

Re:Anandtech

[Anonymous Coward]

They have a large capacitor in the drive. The DRAM is on the ssd and behind the capacitor. If the drive detects a power failure the data in DRAM is written to the ssd memory before the capacitor loses it's charge. This is my understanding.

Cheaper ssd drives may not have the on chip DRAM chip. Research it before you put these in servers. Use the write optimized MLC ssd drives are better geared for logging like Suns ZFS intent log and database logs.

Re:Anandtech

[jhantin]

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?

Write-ahead would be one simple technique. Keep at least a spare block around, and don't blow away the old block until you've copied what you need to keep.

Re:Anandtech

[Anonymous Coward]

They have a large capacitor in the drive.

The X25-M does not.

write optimized MLC ssd

Usually SLC is referred to as write-optimized and MLC is read-optimized.

Re:Anandtech

[DigiShaman]

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.

Good question. I suppose a form of data journaling could be used. Perhaps it's already being implemented in the current firmware. Anyone know for sure?

Re:Anandtech

[Silverfish]

I'm too lazy to go back and reread the entire Anandtech article, but if I remember correctly, it speculates on the amount of memory on Intel's controller and specifically states that Intel doesn't use the controller memory the way you describe, for the exact reason you state. Or perhaps it was the other article they did about the hiccuping drive from... was it OCZ? Either way, I feel pretty confident power loss won't cause data loss (at least not at the fault of the SSD controller)

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:Every SSD has this problem

[afidel]

Perhaps the SLC cells in the X25e are just faster than the controller can cope with but after leaving one writing random data for 24 hours I saw no degradation in speed, if my numbers are right that was about 275 writes to every cell in the SSD assuming a 2:1 flash:usable ratio (.170GB/s*3600s/hr*24 hours*.5). I've also not heard of this kind of severe wear penalty for other SLC based devices like the FusionI/O.

Re:Every SSD has this problem

[AllynM]

The problem Intel fixed is not the same thing you're thinking of. Anand's methodology was flawed, in that he was writing the OS back to the drive in sector-by-sector mode, which is effectively a large sequential write. This acts to heal drives that write combine and is not in line with how that OS would have got there in reality. The subsequent writes he did accomplished nothing more than seeing how far that particular drive could fill the 'holes' in the partition (i.e. how fast it can perform small random writes).

Re:Every SSD has this problem

[AllynM]

It's not how much you write to it, it's how you write to it. Write a bunch of small files to random locations and it will fragment, dropping subsequent write speed. There is a pic of the effect on the last page of my article:

http://www.pcper.com/article.php?aid=691&type=expert&pid=10

What you're thinking of is essentially the response time of the flash itself. Most drives appear to be engineered to assume the flash is at its end of life and keep their timings to that level. No drive I have tested has slowed for this particular reason.

Re:Every SSD has this problem

[afidel]

I was doing 4K random writes to the entire drive using IOMeter, that should be as punishing as anything can possibly be from a small file write perspective.

Re:Every SSD has this problem

[LordKronos]

As AllynM mentioned, this fix addresses a different problem. If you read in that anandtech article, you will see this:

Intel's X25-M: Not So Adaptive Performance?

The Intel drive is in a constant quest to return to peak performance, that's what its controller is designed to do. The drive is constantly cleaning as it goes along to ensure its performance is as high as possible, for as long as possible. A recent PC Perspective investigation unearthed a scenario where the X25-M is unable to recover and is stuck at a significantly lower level of performance until the drive is secure erased once more.

There's not much I can say about the issue other than I've been working with Intel on it very closely and it's not something I'm overly concerned about at this point. I can replicate the PC Perspective data but not by using my machine normally. Through the right combination of benchmarks I can effectively put the drive in a poor performance state that it won't recover from without a secure erase. I should also mention that I can do the same to other drives as well.

Re:Every SSD has this problem

[AllynM]

With 4k writes it was likely writing at a lower speed right off the bat and kept that lower value. Try it at higher queue depths and you will get increased parallelism within the drive. Then you should see higher initial speed that will fall off as the drive fragments.

Re:Every SSD has this problem

[xianthax]

the term "sequential writes" loses meaning in regards to SSD's, a wear leveling algorithm by definition is going to move those blocks around such that they are no longer sequential in nature anyway, assuming that the disk is being used in random nature, that is you do more than just non-stop sequential writes over its life time

Re:Every SSD has this problem

[AllynM]

For write combining SSD's, remapping occurs *within* blocks. You don't necessarily need to write huge sequential files to an X25 to correct this, writes only have to match the block size.

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

[bertok]

I have an OCZ VERTEX 250GB SSD, and it blows mechanical drives out of the water for random IO.

I noticed several reviews that indicated that the Samsung drives do have issues with random IO, but the OCZ drives appear to have no such problems. Yes, you lose performance with random IOs vs sequential IOs, but nowhere near as much as first-gen SSDs. I've seen 6000 random IOPS on a single drive, which is unattainable on anything short of a while tray of disks in a SAN.

I'm not pulling the SSD vs SAN comparison out of my ass, I tested my laptop with the SSD drive head-to-head with the same ~60GB database against two production servers, one with a 20-something spindle SAN volume (shared), and the other with a 3-drive 15K RPM SCSI RAID (dedicated). It won against both for all cases where IO was a significant bottleneck in the query. Obviously, my laptop lost out against the 8-CPU server with 32GB of memory for 'small' queries, but for un-cached data sets, it was usually faster.

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

[Anonymous Coward]

I'm not implying that you are in any way wrong or incorrect, but I recommend that you also read http://www.anandtech.com/printarticle.aspx?i=3531 in order to further reinforce your knowledge and experience in the topic at hand. I am pretty confident that you will find things in that very informative article that you didn't know and that also apply to your situation.

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.

Re:Why'd they call it that?

[X.25]

I also go back to the old x25 datacomm days. so I thought 'huh, what? x25 is coming back??'

what a poor choice of names. clearly, intel has some VERY young people working there that have no idea about past things by the same name.

(admittedly, x25 was not as popular in the US as, say, europe)

o/\o X.25 buddies

Need a bootable CD drive

[tepples]

I thought these SSDs were designed for laptop computers. I read the installation manual, and it didn't give any instructions for what to do if you don't have a CD burner, or if you don't have an optical drive in the computer with the SSD. Or does this update work in UNetbootin?

Re:Need a bootable CD drive

[Anonymous Coward]

they make these things called USB flash drives.......

and they can be made bootable quite easily....

and of course there's always the whole USB optical drive as well.....

Re:Need a bootable CD drive

[tlhIngan]

I thought these SSDs were designed for laptop computers. I read the installation manual, and it didn't give any instructions for what to do if you don't have a CD burner, or if you don't have an optical drive in the computer with the SSD. Or does this update work in UNetbootin?

Actually, Intel's x25 is designed for desktop/server/enterprise use, not laptop use. It's just that the form factor is the same as a 2.5" laptop SATA drive. Mostly because you can fit an affordable amount of flash and the controller in that package. If you went to a 3.5" form factor, you can probably get multiple TB of storage, but affordable it won't be.

If you're designing a laptop with an SSD, you won't go the SATA route. You'd use a spare mini-PCIe slot and use a PCIe SSD (a la the Eee and others). Intel makes a board for mini-PCIe. Saves yourself the cost of all the SATA overhead (connector, power, etc), and since you'll be routing miniPCIe lines anyways...

Re:Need a bootable CD drive

[edmudama]

Actually, the the X25-M is for "mainstream" usage, including laptops and desktops. The X25-E is for extreme workloads, including some server usages.

The X25-M is available both in 1.8" and 2.5" SATA form factors, which are the two most common laptop interfaces today.

PCIe is a bit more limited in a laptop typically, and if you go that route (as a laptop manufacturer) you're generally locking yourself into a single device vendor, since you'll need custom drivers for whichever PCIe board you choose. SATA, on the other hand, allows you to pick any device on the market if you follow the form factor guidelines properly.

Re:Why'd they call it that?

[sortius_nod]

I don't think you understand that term...

http://www.snopes.com/language/notthink/exception.asp

The PR Spin

[symbolset]

Y'know, they contacted the blogger directly, got the actual responsible engineers to listen directly to his concerns, duly investigated and promptly resolved the issue.

Yeah, they're somewhat restrained in their public communications. They're not PR types, they're engineers. That they've been let out of their cave to communicate with an individual member of the community is a big win, especially since they fixed it with a firmware patch. Let's not expect them to host the press conference too. That's too far outside the scope of their skillset.

The fix almost completely eliminates the value proposition of their premium -E line of drives, but they published it anyway. If Marketing were in charge that would not have happened.

I think it's time I bought me a couple of these -M drives.

Re:The PR Spin

[edmudama]

I don't think it necessarily eliminates the value proposition...

The X25-E claims a petabyte of lifetime random writes, plus it's quite a bit faster.

There are applications for each out there, though you're right, the majority of users will be perfectly happy with the X25-M

Re:Why'd they call it that?

[fractoid]

And here was I thinking it was related to the midget autogyro of the same name...

SWIFT No Longer Use X25

[Anonymous Coward]

They use plain standard IP technology, VPN tunnels, LDAP, Certificates, SSH etc on their own network in their own way that makes you wish they stuck with X25.

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

Re:Allyn Malventano Theory - It Can Write Faster!

[socrplayr813]

I agree that it's a bit shady if they are capping like that, but it's up to the competitors to challenge them. If nobody can, then Intel really has earned their advantage and should make as much money off it as the market allows.

Re:Allyn Malventano Theory - It Can Write Faster!

[Anonymous Coward]

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

I think you (and the article author) are indulging in baseless speculation. The graphs don't actually support what you claim. Look at the shape and extent of the valleys below 80 MB/s on the X25-E, compare to X25-M. The -E has wide valleys that go down to ~50 MB/s or so. The -M has narrow spikes that go all the way down to ~10 MB/s. That doesn't look similar to me. And just noting that the graph looks capped at 80 MB/s rather than a "standard waveform" (what is that supposed to mean? who defines the standard for this?) proves nothing... if 80 MB/s truly is the maximum speed at which the media can accept writes, AND the controller is capable of sustaining that throughput, AND the algorithms allow it to do so even in the face of managing fragmentation, the graph looks just like what you'd expect, eh?

The whole idea that Intel needs to cap the X25-M to sell the X25-E is flawed, based on ignorance of what the selling points of the X25-E actually are. The typical customer for the -E could not care less about sequential write performance. The whole point of SSD in enterprise scenarios is transactional performance for databases and similar workloads. SLC is fundamentally better than MLC for this market, due to faster writes, higher reliability, and longer expected lifespan under a heavy write load. (Re: write performance, look up the write latency for the X25-E and X25-M. The -E has much lower write latency, and even though Intel specs both drives at 3300 random write IOPS, benchmarks I've seen have the -E significantly faster on real world random write IOPS workloads.)

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

This argument actually works against your case. Intel's competitors already have a marketing advantage: 200+ MB/s is a nice big shiny number which looks good against Intel's mere 80 MB/s. Most people aren't well informed enough to know that sequential write throughput isn't actually all that important a performance spec for their use patterns, so they assume the latest cheap crap OCZ rebadged must be just as good if not better than Intel's because the numbers are bigger eh?

In other words, Intel should have uncapped by now if they could. They're almost certainly losing sales because of the relatively low write throughput of the drive. Me, I suspect the truth is that Intel's superior random I/O performance (and especially superior random write performance) has a price, and that price is much slower write throughput than the number of channels suggests. Remember, the drives aren't a RAID across the channels internally, and data placement is very important in flash wear leveling.

This is very similar to the situation with processor multiplier locks that they remove in their performance oriented Extreme processor lines.

Multiplier locks are a tool to prevent overclocking. Extreme processors are (A) Intel's way of selling its highest bin sort CPUs, which have to be priced high to regulate demand since they can't make very many and (B) a wink and a nod to overclockers. OCers who want an Extreme CPU for its unlocked multiplier certainly do pay a lot for the privilege, but then again they're well known to be very problematic customers (returning CPUs till they get one that overclocks really high, damaging processors with extreme experiments and making warranty claims, etc.), so perhaps it is appropriate that they pay more.

I don't think this is very similar to the SSD situation, even if one assumes for the sake of argument that there is a cap on the X25-M. The higher performance option being offered is for a completely different market, one which won't buy the X25-M simply because it's MLC.

An easy fix for Intel

[rdebath]

There is an easy fix that Intel may be able to implement in their flash file system. They just have to lookout for free space wipes.

If a block is written with all zeros (all the same byte) they could then free the flash sector and mark the sector as a 'monobyte' in the data structure. The advantage is that it wouldn't take any extra space at write time so the FFS wouldn't get into the state where it's got no space to defragment blocks and so normal ATA commands will be able to get it out of it's stuck slow state.

TRIM without waiting for a standards committee.

It might not be possible for the hardware but it could give more stable write rates and some insane read rates if they did!

PS: Yes, Flash file system is right, they have implemented an FFS on their drive that provides access for the user to write to one, large, file. Like a partition table that can be looked at as a very minimalist filesystem too, some partition tables even have names for the partitions.

Re:Why'd they call it that?

[GordonCopestake]

Perhaps they should call have called it X26 then? :p

A first too

[Ilgaz]

Intel could be the first among vendor/OS developers to admit drive fragmentation COULD BE an issue, in certain usage patterns. MS themselves kinda admit too but as you see the negative feedback about NTFS as result of it, I think they may slowly back down from suggesting it to users.

As a guy in video business, you can't believe how much we are blamed, called stupid, old fashion, not reading OS documents when one sees we defrag drives in certain cases. Windows, OS X, Linux won't really matter. When one half of half terabyte of file is beginning drive and the other half is at end of drive, "fps" will drop. (SAS) SCSI, 10K, fiber won't matter. I bet audio guys working with gigantic files and realtime must be getting similar feedback too.

What bothers me is, for some reason, as it is a mostly electronic issue, OS vendors themselves feel like they are being blamed. Even Apple who does a lot to prevent fragmentation, defragmenting on the fly (based on stability comes first rule of course), using once mainframe technology like "hot band" doesn't make it clear.

I once got laughed at for saying it, "defrag" is a luxury system performance enhancement. You can live without it of course but sometimes, it may really matter.

If I had SSD (I plan 32GB for startup drive), I would at least optimise its HFS+ B-Trees and proactively prevent issues with a commercial disk utility like Disk Warrior. I don't think journaling would be good for it anyway and I don't need "hot band" stuff which would happen if journaling enabled.

Re:A first too

[phayes]

Intel could be the first among vendor/OS developers to admit drive fragmentation COULD BE an issue, in certain usage patterns.

Not quite, AFAIK, Anandtech broke the story here and though he did say the Intel was the SSD vendor who was the least affected by the fragmentation bug, he also details that OCZ had already made great progress in resolving it's issues and becoming the SSD price/preformance king.

Indicative of a much larger problem

[EmagGeek]

Engineering Culture worldwide has been completely usurped by the marketing and quick-buck executive paradigm. I've been working in engineering for a decade and the notion that a product should work properly before it is released has been thrown out the window and splattered on the street.

"Get it out the door, and worry about issues later," is the mantra. Final Product Release has become the new Beta Test phase. One look at GMail and you'll see what I am talking about. GMail has been in "Beta" for what, 5 years now? And they're charging people?

The iPhone is another prime example. The first release was just a Beta or dare I say Proof of Concept. It didn't even support 3G and had mountains of issues.

The list of example goes on and on and on.

Re:Indicative of a much larger problem

[Courageous]

You're mistaken that it's really the fault of "quick-buck executives". It's the market, and the people in it.

My wife often complains to me when some bit of software is manifesting a bug. She asks "why can't they make software that just works?!"

The answer is, "they can, but you wouldn't buy it". Bug free software is quite expensive. The programmers that write in bug free environments are typically 4-8X less productive, on a line count basis, than programmers who work in non bug free environments.

Unless such things are essential to the market in which they are used (flight control systems, financial transaction systems, etc), the market simply will not bear the costs of such development.

What I am trying to say here is that there are significant market forces at work that reward these "quick buck executives". Basically, they are making what the people want. No, kidding.

C//

Pace of progress

[DragonHawk]

"Something tells me the SSD scene is moving so fast that within literally 6 months one of these 2 companies (or a competitor taking note) will have a product superior in size, speed and price to those 2 very very soon."

And this is different from the rest of the computer hardware world how? :) Everything is always getting bigger, faster, cheaper, smaller, whatever.

One thing I've learned is that, in general, one should decide on a budget and make your purchase based on what's available today. Something better is *always* coming down the pike. :)

Re:Pace of progress

[n1ckml007]

Compared to existing technology (magnetic platters), the $/GB is still quite high.

Re:Pace of progress

[AbRASiON]

Unlike processors, memory, DVD drives, monitors or the vast majority of parts in a computer, SSD's are just breaking the infant stage of introduction.
This is why you see an Intel X25M at 700$ 5 months ago, being 400$ now.

The whole industry moves fast but the next 2 years SSD's will be catching up to the more 'reasonable' pace of hardware today.
Therefore it's prudent to probably wait only a small amount of time for huge increases.

Re:Pace of progress

[maxume]

Most people only 'want' flash drives at this point, and there are improvements being made that seem to be in addition to the normal doubling cycle (i.e., real world experience is still getting fed into the basic designs of SSDs). So while I agree with you that picking a budget and then buying is the correct approach, the SSD market looks like it will be much better in 12 months, out of line with predictions based on a quick look at the current market.

Analogy.....

[IHC Navistar]

Fragmentation problems?

It's like playing "52 Card Pick Up", except instead of using 52 cards, you're using 80 BILLION bits!

Current state vs delta

[DragonHawk]

"Compared to existing technology (magnetic platters), the $/GB is still quite high."

Um... so it is, but it seems to me that your statement is kind of a non sequitur. My post is all about the pace of change, the the slope, how fast things change from "new" to "old". Not the current state.

Re:Current state vs delta

[Sj0]

Computer technology is progressing FAR more slowly than it used to.

In 1991, you'd be using a 386 or 486 with 2-4MB of RAM. By 1999, you'd be using an Athlon with at least 512MB of RAM. That's a MASSIVE difference.

By contrast, in 2001, you'd be using a Pentium 4 at around 2GHz with about a gig of RAM. Today, you could be using the same machine. Sure, there are some cool technologies that have come out since then -- 64 bit processors are ubiquitous, and multi-core technology is insane, but we're not looking at anything like the difference between 1991 and 1999.

Re:Current state vs delta

[n1ckml007]

That's a good point I have a P4 3.2Ghz that is still used as my main desktop at home. The MOBO is an Abit from 2004, and it runs XP SP3 nicely.

That's what I asked

[tepples]

Anonymous Coward wore out the keyboard's period key:

they make these things called USB flash drives.......

and they can be made bootable quite easily....

That's why I asked whether or not this CD image was compatible with UNetbootin, a tool to copy some operating system installer CD images to USB flash drives.

and of course there's always the whole USB optical drive as well.....

Two problems:

1. *Sold separately. Or do you claim that anyone who can afford an X25-M drive can also afford a USB burner?
2. Some BIOS versions that can boot from ATA or SATA optical drives and from USB mass storage devices cannot boot from USB optical drives. But is this true only of older machines, not Eee and later?

Motherboard SKUs

[tepples]

If you're designing a laptop with an SSD, you won't go the SATA route. You'd use a spare mini-PCIe slot and use a PCIe SSD (a la the Eee and others). Intel makes a board for mini-PCIe. Saves yourself the cost of all the SATA overhead (connector, power, etc)

Unless you want to have one motherboard for both hard disk and SSD versions of a product. Or do hard disks come in mini-PCIe now? I thought that went out with hardcards back in the mid-1980s.

Re:Definitely

[colinnwn]

I'm in a similar situation now. My parents own a Samsung DVD-R120 that plays but won't record DVDs. I thought they were doing something wrong and didn't get around to checking it until this weekend 3 years after they bought it. I discover on the internet many people have the same problem with the AXAA submodel, but almost no one reports the problem with the XAA submodel.

I'm so frustrated that Samsung didn't proactively contact customers with this device, or at least post a notice to their support forums about the problem and offer a resolution, that despite how much I love Samsung displays I am considering swearing off Samsung products in general. We also had a cheapo Samsung DVD player that died just out of warranty, and an expensive upconverting DVD player that works ok but has a laggy UI. And it frustrates me how all the DVD and TV remote controls from the previous generation of devices have to be so precisely aimed.

Re:The alternative

[colinnwn]

Do you have more evidence besides this IBM situation people drop the brand on early reports of proactively resolved problem products?

For me, if a company publically says a particular product may have problems, but we will support it to the point we can and we'll double the warranty, I will be very likely to stay with them as a customer on other products. And I might consider the problem product.

Now if a product gets some widely reported negative publicity on problems that may occur on a small portion of units (yet as a customer it is hard to verify that), and I don't feel like the company has done enough to support that customer, identify the problem, and alleviate concerns in general, I'll be very unlikely to remain a customer of theirs.

In the end I guess companies just decide whether to throw the dice on a big bet, or walk away with a small loss.

Re:Attention linux users

[Anonymous Coward]

Well, that will probably give a "permission denied" as the stdout forwarding is done by the normal user. :D

If you want to nuke a disk then the command would have to be sudo sh -c "yes > /dev/sda".

X25-M + Bootcamp

[zurmikopa]

I know the X25-M didn't work with bootcamp before, does anybody know if the firmware update also addresses this issue?

Re:X25-M + Bootcamp

[LordKronos]

Couldn't be bothered to read the article? Check out page 7. It covers it in detail, but
1) no it doesn't fix anything
2) not all drives had problems in Bootcamp to begin with

Re:X25-M + Bootcamp

[zurmikopa]

Nope, only had a minute or two free at the time and forgot about it later. (The article is kind of long)

I was hoping some kind soul like yourself would have read the article and possibly others, see my question and answer it.

Thanks :)

These aren't the Droids your looking for.

[rossy]

Move Along...