MMM
Results 1 to 13 of 13

Thread: SDD degradation

  1. #1
    Xtreme Mentor
    Join Date
    Feb 2009
    Posts
    2,597

    SDD degradation

    This post is to try and work out what the affects are of SDD degradation. To kick it off:

    An SDD that shows 50% of written data is not showing how may cells have written data. In addition technology is being used to limit wear by distributing data across cells to limit the times they are written......if you try to shrink a SDD partition you can't do it.

    Wear technology and the lack of ability to flush the cells of unwanted data. I'm still trying to get my head round this so please bear with me.......

    If you have a 32GB SDD drive and during the Win 7/ Vista install process temporary files create 24/26 GB of written data the cells are technically 80% full as the files that were temporarily written are still kept on the SDD cells.

    Does that not mean that the SDD is already 80% degraded by just installing the OS? Add a few basic programs and take into account temporary files being written in the process and the drive's cells are likely to be 100% full, so why would a 32GB drive benchmark well on a fresh install and then see a significant drop in performance after a few days?

    From what I understand it does not matter how much data is in the cell, it takes the same time to overwrite it. So once the cells are full the drop in performance should be definable based on how many cells are being overwritten.... but is it?

    The other thing that is puzzling me...in the scenario of a 32GB drive OS install leaving all cells with written data, how does the drive know where to distribute data to prevent wear? Assuming it does know how to overwrite data on cells that are not required why can't a process be developed to clear data that is not wanted? If wear distribution can only work on cell with no data does the drive wear out quicker as a result?


    I guess what I am getting at is this.

    1. Does degradation affect real use performance? (I can answer this at least for the X25-E. No).

    2. If the time taken to overwrite a cell is quantifiable the drop in performance should match depending on how much data is being overwritten. (If it isn't that would seem to raise more questions).

    3. Does degradation occur quicker on smaller drivers in comparison to larger drives? (If it doesn't that would seem to raise more questions).

    4. Does degradation affect only read/ write speeds or does it affect IOPS as well?
    Last edited by Ao1; 03-10-2009 at 01:15 AM. Reason: Item 4 added

  2. #2
    Xtreme Member
    Join Date
    Jun 2005
    Location
    United Kingdom of Great Britain and Northern Ireland
    Posts
    464
    I think where mechanical drives have fragmentation, SSDs have [I don't know of a good name for it] degradation.

    Something that seems to have worked from OCZ users is to use HddErase 3.1. This seems to put the cells back to an empty state. The software is designed to be used to wipe out data on a drive for security purposes. Another piece of software with the same purpose is Killdisk, however that has the opposite effect, and instead of restoring performance it severely impairs it. The thread is here.

    I suspect this issue is one that would be solved by a process similar in function to defragging a mechanical drive. The program will simply consolidate free space, and erase the free space on the drive to make sure it's ready to receive data without being erased at that point in time.

    It's possible that Diskeeper's Hyperfast already does this. I have read on the OCZ forum that a user found benchmarks did not give consistant results unless this was run first.

    As for the questions.

    1) Have you been using the X25-E for long? How much of the capacity have you filled?

    By what you have said above the drive will write to the least written parts of the drive first in order to ensure reasonable wear levelling. By that then I think it's reasonable to say that the more free space available, the longer it would take for this to get "filled up". Conversely on a drive with very little free space either some kind of operation would need to be regularly performed to clean up the disk, or performance would quickly degrade. I believe that this has been anecdotally reported by Intel users.

    What I suspect is that the Intel drives will function just fine. Perhaps a little degradation that only a benchmark would pick up will be experienced over time. However eventually the performance would significantly degrade, and would require some kind of maintenance. Just when this would occur is a mystery, however given that I've yet to see an Intel user here report that their drive has hit the skids then I think it's reasonable to say that the drives are perfectly functional, and degradation doesn't affect real performance, at least in the short to medium term.

    2

    As I understand flash drives have erase block sizes much larger than the blocks they store data in. To illustrate, let's assume it's 512 kilobytes. A piece of data 4 kilobytes in data needs to be written. An erase cycle needs to be done, and then a write cycle. This means the total amount written is 516 kilobytes. Even if a drive has been completely filled an erase cycle will not need to be done each time a 4 kilobyte file needs to be written so performance will seemingly randomly vary, and on a sustained benchmark it will mean up to 50% slower write speeds. If the data being written is larger than 4 kilobytes then the impact on performance will be lesser, since the overhead from erasing will be smaller in comparison to the data being written.

    3

    I think I covered that above. On a pristine 30GB drive with 50% capacity written the drive will fill up faster than a 120GB drive with 50% capacity written, simply because there's 15GB that needs to be written before erases need done on one drive and 60GB on the other.

    A disclaimer : I'm a little out of my depth here, this is just what I -think- is right, but I certainly wouldn't stake any money on it.

  3. #3
    Xtreme Mentor
    Join Date
    Feb 2009
    Posts
    2,597
    The drive has been running at 50% capacity of valid data between January to the beginning of March. No perceivable slow down experienced in that period. The drives were formatted recently and benchmarks showed a subsequent improvement. http://www.xtremesystems.org/forums/...=219707&page=2

  4. #4
    Xtreme Enthusiast
    Join Date
    Feb 2009
    Location
    Colorado Springs, CO
    Posts
    957
    I would not recommend installing Vista on any drive smaller than 60GB. The reason I say this, is even if your base install fits, as you use the machine, the WinSxS folder gets much larger as you install/uninstall programs. On my current machine the WinSxS folder is nearly 20GB alone, not counting any swap file, programs or Windows on that drive.

    After a few months of running on a 30Gb drive, you will start to see that no matter what you do to clean up temp files, you will be constantly plagued by out of space issues.

    Oh, and Audience... it is SSD, not SDD

  5. #5
    Xtreme Mentor
    Join Date
    Feb 2009
    Posts
    2,597
    Quote Originally Posted by 3Z3VH View Post
    .

    Oh, and Audience... it is SSD, not SDD
    Oh well could have been worse...... I could have called it STD

  6. #6
    Xtreme Member
    Join Date
    Feb 2009
    Posts
    265
    I just got myself samsung 64gb ssd, will that be big enough ??

  7. #7
    Xtreme Enthusiast
    Join Date
    Feb 2009
    Location
    Colorado Springs, CO
    Posts
    957
    Should be enough, as long as you are frugal with what programs you install on C:\. If you install any apps that don't need fast disk access to program files, I would suggest installing them on your secondary storage drive.

  8. #8
    Xtreme Member
    Join Date
    Feb 2009
    Posts
    265
    Quote Originally Posted by 3Z3VH View Post
    Should be enough, as long as you are frugal with what programs you install on C:\. If you install any apps that don't need fast disk access to program files, I would suggest installing them on your secondary storage drive.
    Cheers Have been reading up a bit on it and people are moving their page file systems onto their normal HDD's and moving programs cache also onto good old HDD's. Wonder whats the opinion of people on here at doing such thing

  9. #9
    Xtreme Enthusiast
    Join Date
    Feb 2009
    Location
    Colorado Springs, CO
    Posts
    957
    If your SSD is not subject to huge Write lag or degradation, I would keep the swap and cache on the SSD. The added speed of the SSD would help in both of those situations. Only reason to move them off a good write performance SSD is to save room.

  10. #10
    Xtreme Member
    Join Date
    Feb 2009
    Posts
    265

  11. #11
    Registered User
    Join Date
    Oct 2004
    Location
    Portugal
    Posts
    48
    Quote Originally Posted by oggie View Post
    Can you please tell us something about their performance?

    EVGA X58; I7 920 C0; OCZ Reaper 1866 88824; Sapphire 5870; 1x SSD; 2xWD 1Tb GP + 1 Tb WD Black; X-FI Titanium; Corsair HX 620; Lian Li PC-50P

  12. #12
    Xtreme Member
    Join Date
    Feb 2009
    Posts
    265
    Quote Originally Posted by gonx_me View Post
    Can you please tell us something about their performance?
    I have not actually recieved it yet, just made order this morning. Here is little thread http://forum.novatech.co.uk/showthread.php?t=7285

  13. #13
    Registered User
    Join Date
    Oct 2004
    Location
    Portugal
    Posts
    48
    Thank's. I'm going to read something about it. I want 2 for raid 0!

    EVGA X58; I7 920 C0; OCZ Reaper 1866 88824; Sapphire 5870; 1x SSD; 2xWD 1Tb GP + 1 Tb WD Black; X-FI Titanium; Corsair HX 620; Lian Li PC-50P

Bookmarks

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •