Go To Latest Video ReviewsIntel Updates SSD Firmware Comments
| Author | Gordon Kelly |
| Published | 14th Apr 2009 |
Comments for Intel Updates SSD Firmware
Keith said on 14th April 2009
Gordon said on 14th April 2009
@Keith no, SSDs operate completely differently than regular HDDs but drive performance on all SSDs has been shown to drop off as the drive gets fragmented over time.
smc8788 said on 14th April 2009
SATA is just a storage interface, so although SSDs still use the SATA interface, it has no bearing on how the drive operates. I was under the impression, however, that SSDs didn't need to be defragmented (hence no fragmentation), but I guess I am wrong? Surely with their incredibly low random seek times the performance drop off would be minimal?
BinnsY said on 14th April 2009
Can you not just use standard defrag programs to defrag an SSD every so often?
Gordon said on 14th April 2009
We all thought so at the time, but fragmentation does still cause problems. Problem is defragmentation isn't recommended in general as it causes a lot of wear on the SSD cells. Hence the fix...
BinnsY said on 14th April 2009
Ahh of course, forgot about the limited number of write cycles, makes more sense now.
Keith said on 14th April 2009
mmm, for some strange reason I believed SSD's didn't have a FileSystem, the reason why you can install NTFS, FAT, Ext2, RaiserFS etc onto them.. IOW: It's the OS responsibility for the FileSystem, are you 100% sure this uses a filesystem, and not some sort of block/sector mapping like I pointed out?. Maybe it's some sort of mapping to reduce the Read/Write cycles used.
Keith said on 14th April 2009
@smc8788: SATA is just a storage interface, so although SSDs still use the SATA interface, it has no bearing on how the drive operates
Yes, I know that, the same is true for PATA. But that has nothing to do with what I was talking about. :) My point was that the term FileSystem was used, and I believe that is not the case, maybe the term FileSystem was used as it sounds better than block manager.
Maybe this link will explain things better than me. ->
http://www.pcper.com/article.php?aid=669
FuricTrax said on 14th April 2009
Anandtech.com has a brilliant and perhaps overly in depth article about SSDs and why the performance degrades so dramatically when they start to get full. Continual optimization by the drive can help massively to reduce this performance hit which for some drives can be a hit of 75% or more in certain scenarios. See here
http://www.anandtech.com/storage/showdoc.aspx?i=3531
Dreamwalker said on 14th April 2009
PC Perspective has an article which actualy "caused" the new Intel firmware. Recommendet reading: http://www.pcper.com/article.php?aid=691
Add Your Comment
Add your comment
You must be logged in to comment. Login or register here.
@Despite this modest description, the primary focus of the firmware is to fix issues regarding long term file fragmentation
Fragmentation?, I thought these devices acted like normal SATA drives. IOW: It's the OS that's responsible file sector allocations, and SATA drives know nothing about files, sectors maybe. Could it be some sort of Sector mapping there talking about here, a bit like the bad sector mappings you get with normal HD's?.