Announcement

Collapse
No announcement yet.

Performance degradation, refresh tool?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Re: Performance degradation, refresh tool?

    Originally posted by UltraTux View Post
    "as clean"? That's the silliest name ever. Can't even google it. Does it work on Snow Leopard? U have the link?
    "As cleaner" was written and submitted by an OCZ Forum member. All it does is write FF to the remaing space on the SSD *after* the data has been consolidated to the front of the SSD using PerfectDisk (a defragmenter).

    There would surely be some tool for Snow Leopard that can allow you to create a file with "FF" hex data to fill the free space on your SSD, then delete the file upon it's completion?
    DIY ViDock O/C HP dualIDA | PLL pinmod | grub2 loader Storage 1.8" ZIF SSDs: ProIV, Photofast V4 | ODD caddy SSD/HDD | e-sata/usb

    Now: 12" HP 2510P [email protected] 2GB 64GB ProIV SSD + 500GB.caddy + GTX470/3dmark.vant=P5561 | Win7/64 XP.sp3 Ubuntu9.10
    Now: 14" Compaq CQ41 i5-430M 2.26 4GB 500GB + GTX470@x2/3dmark.vant=P10481 | Win7/64
    Next(Q1.2011): Sandy Bridge + pci-e 3.0 + GTX470-OC/3dmark06=22k

    Comment


    • #17
      Re: Performance degradation, refresh tool?

      To add to what nando said about defrag and consolidation, for me the main reason to consolidate free space is the following:

      As we know, writing to clean flash is fastest and writing to dirty flash is termed as perf. degradation. When restoring performance with wiper, if you consolidate before then you will be able to have the max number of flash blocks that is physically possible cleaned. This is because the drive must erase in 512KB blocks at a time. Regardless of if there is 512KB of user data in the block or something like 1KB (any number less than the full capacity) for random writes in most cases this "slack" effect will happen where the next file will be written to another block.

      When consolidating, the program will group together a file or files so that as many blocks as possible are completely used, meaning more blocks are free. To simplify this example, lets pretend that 1GB is 1000MB just for a sec. You can have a Gigabyte of datastored in for example 5500 blocks. After consolidation, this will become much closer to the ideal of 2000 blocks (2 blocks per MB, times 1000 = 2000) after doing this you now have an extra 3500 blocks that can be cleaned up by wiper.exe or AS-Cleaner, meaning that your drive will go for longer before perf. degradation kicks in again.

      Tony Trim is to be used bearing in mind that it does take up a couple of write cycles to do, meaning it minutely impacts on drive life. Also, some users have noted a decrease in performance using it on an Indilinx drive. It's also IMO not recommended on some other SSD's with different controllers. Cases of its use on Samsung drives has been posative but I've used it on my X25-M with the oposite effect to the one desired. With my X25-M there's a typical pattern to benches when it's got no clean flash left. Random 4K Write clearly drops by a fixed amount each time. I tested the Tony Trim method first on a degraded drive and it did nothing to improve performance. I then cleaned it with HDDErase, re-imaged and verified it was back to peak performance then did a Tony TRIM and it degraded the drive as per the typical speed decrease I mentioned.

      AS-Cleaner is a useful tool in certain situations. Also Diskeepers "HyperFast" (for SSD's) is very effective. It also works on my X25, restoring 4K random Write to ~90% of fresh instead of the degraded 80% of fresh (X25's don't really loose much performance TBH, they're amazing in that respect).

      As for the wiper program, I may be missing something but as far as I can tell it will work with any Indilinx drive. There shouldn't really be any issues. If the issues you encountered were slow performance in AHCI and RAID mode then this is a common occurance and has never been fixed by Indilinx. A full wipe in IDE mode takes around 60-90 seconds where as in AHCI it takes (for me) about a hour. That's why I've gone back to Legacy IDE, even though AHCI gives my other SSD (X25-M) a good near 8% boost in reads.


      ****EDIT****
      I'm sure it would be possible to create a program to write FF (logical 1's) to the SSD in OS-X however I'm fairly sure one does not currently exist. This is similar to the Windows situation, as after all, before AS-Cleaner was modified on OCZ's Tony's request to handle "FF" writing, a tool wasn't available for Windows either. Maybe there will be someone reading this with OS-X skills sufficient to create such a program, that'd be cool.
      Last edited by Psycho101; 12-14-2009, 05:35 PM.
      Coolermaster CM 690 II advance Case
      Corsair HX750 (CWT, 91%(80+ Gold rated @230V) single 62A 12V rail
      P55A-UD4 v2.0 @ F14
      Core i5 760 @ 20 x 201, 4.02GHz
      TRUE Black with a single Noctua NF-P12 pumping out 55 CFM @ 19db .
      2 x 2GB Mushkin Ridgeback (996902), @ 7-10-8-27, 2010-DDR, 1.66v
      2 x Gigabyte GTX 460 1024MB in SLI (Pre OC'd to 715MHz core and 1800MHz VRAM) @ 850 Core / 4100 Mem.
      Intel X25-M Boot Drive (OS and Programs) 200MB/s Read & 90MB/s Write
      Corsair X32 200MB/s Read & 100MB/s Write
      WD Caviar Blue 640GB C (Steam, Games, Storage, Temp Files & Folders, etc)
      Samsung F3 500GB Backup/Images
      Noctua 1300RPM 19dB case fan (rear extraction)
      3 x 140 MM Coolermaster LED fans (one front intake, one top extraction, one side intake)
      Dell Ultra Sharp 2209WAf E-IPS @ 1680x1050

      Comment

      Working...
      X