Bad news about the Intel X-25M dropping in performance very quickly...

[barkmonster]

http://www.pcper.com/article.php?aid=669&type=expert&pid=1

Here's a few quotes from the article above:-

Since the flash locations are isolated from the OS by a lookup table, these fragments can not (presently) be directly manipulated by the OS, leaving any defragmentation efforts to the SSD itself.

Once internal fragmentation reached an arbitrary threshold (somewhere around 40 MB/sec average write speed), the drive would seem to just give up on ‘adapting’ its way back to solid performance... In several tests our write speeds dropped to 25-30 MB/s and simply refused to recover on their own

Even with AHCI significantly boosting the reading of fragmented areas, the X25-M can still fall prey to its own fragmentation. The point to take home is that merely placing an OS or any other group of files onto the Intel drive is enough to bring it below its specifications.

It ends with this...

There Is No Defrag (yet)... All is not lost, however, as the ATA spec is being updated to include special commands such as “TRIM”, “DISCARD”, and “UNMAP” (a SCSI command).

Here are some things to avoid:

* Heavy temporary file activity (think temporary internet files).
* Heavy page / swap activity.

So that makes them useless as the "turbo" system drive everyone is buying (or wanting to buy) once the drive fragments itself at present.

 

[kraftzwerg]

That doesn't sound good ... I was contemplating to get one of those ...

Is this a problem that will affect all speedy SSD drives, e.g. the new Samsung 256 GB one as well?

 

[barkmonster]

From the article it sounds like it's just going to be an OS tweak and a firmware update to the drives and it's problem solved but I suppose at present it's the fact none of those updates are available.

I read through the whole article and it's to do with the very large block size used by SSDs. It's 512Kb which is huge.

I suppose the mac file system wasn't much better than that with large drives before we got HFS+, regardless of how big a drive was, the OS could only split it into 65,536 different allocation blocks, wasting a fair amount of space.

It's the fact SSDs will bunch together smaller files to fill those 512Kb blocks that causes the slowdown, delete a few of them and bits of files start being spread all over the place.

It was the fact sites like that were the first to point out the fact that de-fragmenting was meant to be a thing of the past with SSDs and now it seems to be their biggest weakness that is such a shame.

 

[Amdahl]

This doens't affect Mac...

... because everybody knows OS X doesn't fragment. And when it does fragment, it defragments itself by using Time Machine to go back in time and prevent the fragmentation from occuring. Steve Jobs wrote the algorithms himself to do this. In Objective-C. All Cocoa.

He won an Emmy for it.

 

[Schtumple]

... because everybody knows OS X doesn't fragment. And when it does fragment, it defragments itself by using Time Machine to go back in time and prevent the fragmentation from occuring. Steve Jobs wrote the algorithms himself to do this. In Objective-C. All Cocoa.

He won an Emmy for it.

In an elevator, which raised concerns about the stories legitimacy, but then the newbie poster promised they were close with apple, and everyone believed it...

I don't see how this is such a big news story, you'd have to really hammer it to notice any slow down, and by the time you did notice a big slow down, Intel would've released a firmware update...

 

[Abstract]

Wirelessly posted (Nokia 5800 Tube XpressMusic : Mozilla/5.0 (SymbianOS/9.4; U; Series60/5.0 Nokia5800d-1/10.0.010; Profile/MIDP-2.1 Configuration/CLDC-1.1 ) AppleWebKit/413 (KHTML, like Gecko) Safari/413)

I don't get the first quote in barkmonster's post. How could an SSD be responsible for defragmenting itself? Once it defrags itself, the OS won't know where the files are afterwards because it never got the new addresses, since it wasn't in charge of the process. I'm sure there's just something I don't understand.....

 

[Catfish_Man]

I don't get the first quote in barkmonster's post. How could an SSD be responsible for defragmenting itself? Once it defrags itself, the OS won't know where the files are afterwards because it never got the new addresses, since it wasn't in charge of the process. I'm sure there's just something I don't understand.....

This isn't about filesystem fragmentation, it's about the erase blocks on the SSD. It will still present it to the outside world (the OS or disk controller in this case) as though things were in the same place.

 

[alphaod]

Diskeeper 2009 can defragment SSDs.

 

[barkmonster]

Diskeeper 2009 can defragment SSDs.

not according this post on the OCZ site.

Just completed two manual defrags (just to be sure). There was some (minor) improvement in performance (I did a couple of small file copy benchmarks). However, the defrag did not solve the stuttering at all.

On my system, Diskeeper 2009 Professional with Hyperfast did unfortunately not result in a notable improvement. And performance was nowhere near SteadyState or MFT. Others might have better luck than me though, so do not let me discourage you from trying it out and reporting the results here.

http://www.ocztechnologyforum.com/forum/showthread.php?t=44422

Although it is one of those cheap SSDs brands that stutters etc... so there could still be a "get what you pay for" factor but it doesn't help that every utility that claims to help SSDs is not only Windows only but plenty of people dispute the claims too.

 

[dwsolberg]

Remember that the Intel hard drive is designed to adapt to use. In other words, it observes what you access and write, and prioritizes that. I think these reviewers were constantly installing and reinstalling software and running benchmarks and rerunning benchmarks and trying to defragment and so forth. In other words, they were likely making the algorithms become unstable by constantly changing usage patterns how they were pushing it.

In real life, I haven't seen any reviewers on amazon.com or newegg.com complaining about changes in speed over time. Most of the problems there (and the average review is 5/5 and 4.5/5 respectively) are about arcane Windows settings and other things that aren't necessary with OS X.

In other words, I'm not sure there is any problem for everyday Mac users. That said, I'm not buying one of these drive quite yet, even though I want to.