Defrag SSD ?

[austhinker]

Intel clearly recommends NOT to defragment their SSD. You have had many answers telling you in great detail why this is so. If you are so compulsive about defragging that you can't help doing it to your SSD you would do better to see a shrink than to defrag your precious SSD and ruin it fast

With the price of SSDs dropping, and the price of shrinks, it's probably cheaper to ruin a few SSds, especially as you'd need more than one visit to the shrink!

Seriously though, the sugestion of a once or twice a year clean-up seems like it wouldn't do too much damage, although it's probably better to backup your SSD to another drive, wipe all pages, and restore, rather than running a defrag program which would probably write many pages several times.

It'd be interesting to see how a backup, wipe and restore affected performance of a heavily fragmented drive!

 

[austhinker]

I seem to recall that when NTFS first came out they said it didn't need defragmenting!

Maybe in a couple of years SSDs will be designed for SSD friendly defrag programs to defragment them, with dialog between the program and the wear-levelling firmware. One option is for the data to stay in the same locations, but the pointers to be changed so the file system sees the pages as contiguous - the flip side of wear levelling, where the data moves but the file system sees it as being in the same place.

Who knows? Maybe they'll even come out with self-defragging drives with file-system aware firmware!

 

[whittled my own computer]

Another possibility, IF one is convinced there is a problem in the first place, is to clone your SSD to another drive, then clone it back again.
XXClone and other similar programs seem to effectively defragment automatically during the process on HDDs, but not as a separate function per se. They simply are optimizing the writing, without a forced "defrag."
Since it appears quality SSDs are actually distributing writes across the chips at all times the concept may all be moot. The issue really boils down to the fundamental difference in how bits are written and retrieved, versus a spinning platter.
That does raise the obvious related question, though - if one uses Acronis or another backup type software that is constantly re-writing in a special "partition" (not even sure if such is appropriate for SSDs), would that not also represent a lot of accelerated processing on the SSD, compared to just routine computing?
Even though good manufacturers assure us that they are solid, I expect as SSDs prevail the sheer numbers will more quickly reveal patterns of failure or problems noone could have foreseen. Beta testing isn't the same as real-world experience. By then, new maintenance software completely different in nature, but analogous to "defraging" today, will probably appear, requiring us to run something entirely new to fix entirely different (but achingly familiar) problems.

 

[yahoojusnit]

It is hard to say whether it is necessary or not. File fragmentation will always affect performance no matter how data is written. Windows 7 was developed in 2008 when SSD development industry-wide was still in its infancy.

Windows 8 will be a better measuring stick since Windows Defragmenter is indeed configured by default in Windows 7. And I don't mean the preview, Alpha or Beta.

IMHO, Diskeeper is worth every penny. If you have already shelled out the cash for a SSD, why are you pinching pennies anyway?

 

[badak]

When INTEL said no - did they have an ulterior motive?

 

[dhitson]

The seek delay associated with head movement in legacy drives is eliminated in SSDs. However, the more fragments a file has, the greater chance there is for logical errors - either in the file itself or in the MFT (assuming yopu're running Windows). Considering that a "read" of a single file fragment takes a finite amount of time, any system still has to read a subsequent fragment after its preceding fragment. The more fragments - the greater read (/write) time rule still applies.

If an SSD is installed in a system designed to maximize the throughput, then an equivalently fragmented comparison between an SSD vs. legacy will show a significant performance gain for the SSD. However, interfacing SSDs with older systems often yeilds less than expected performance results.

SSD performance is still nowhere near that of RAM, but systems optimized for SSDs can bring a smile to a tecchy's face!

Just keep in mind - SSDs primarily address hardware performance issues - not logical issues.

Also, read the user reviews of specific brands before purchasing - some of them are quite unreliable and will embitter your technological taste buds.

 

[rugman1969]

I hate to bring this up again, but I just defragged my Samsung ssd. Programs has started running and booting faster again, including boot up speed. I don't know how much life I may have destroyed by doing this, but it definitely improved performance. My drive is a bit over a year old, and this is the first time I have defragged. It took 10~15 mins. Just my input on this topic.

 

[magicpotato]

I "defragged" my nexus 10 tablet using android defragger. Ran noticeably faster after the defrag though I found just a simple reboot increases performance tremendously. Don't know how much damage I did but it seems to be running better regardless. It's also hard to tell if the apps are actually defragging or just "trimming" (which I guess could be the same thing but it would be nice to have some clarification.)