Intro Download and install Frequently Asked Questions Tips and tricks

Homepage







© J.C. Kessels 2009
MyDefrag Forum
April 20, 2014, 02:19:01 pm *
Welcome, Guest. Please login or register.

Login with username, password and session length
News:
 
   Home   Help Search Login Register  
Pages: 1 2 [3]
  Print  
Author Topic: Thanx and request  (Read 13870 times)
JDPower
JkDefrag Hero
*****
Posts: 207


View Profile
« Reply #30 on: October 21, 2007, 05:24:31 pm »

Quote from: "Lundholm"
I think that application files are also moved as part of the XP optimization.
Sorry but they're not.
Quote from: "Robert_S"
So if you donīt test you will see it only in one way.

Where did I say I'd not tested it. I ran without prefetch once before by accident after breaking it and applications were launching noticeably slower. Enough for me to realise something was wrong and eventually pinpoint it to prefetch being broken.

I have also tried disabling prefetch completely for the supposed performance improvement that some claim and noticed no improvement at all.

The issue of prefetch always causes arguments because everyone thinks their way is the right way and there's not enough information from MS to prove either way. So I'm not gonna carry on arguing my point here, what I do works for me and I know it works for me. I don't really care if anyone else disagrees, if what you are doing works for you then keep doing it and be happy :roll:
Logged
jeroen
Administrator
JkDefrag Hero
*****
Posts: 7182



View Profile WWW
« Reply #31 on: October 21, 2007, 10:37:18 pm »

Quote from: "Lundholm"
Yes, but this is exactly what the option 9 sorting does. Files that are used at exactly the same time, will be placed next to each other during the sorting.

The "-a 9" optimization (sort by last access date) cannot always be used. Vista has the NtfsDisableLastAccessUpdate registry key set on by default, and on other Windows versions the user may have turned it on. And then there are programs such as virus scanners and backup programs that will reset the last access time to "now" for all files on disk.

Secondly, the "-a 9" optimization will sort the last accessed files to the begin of the disk. If a program uses 2 files when it starts up, then the second file will be placed on disk before the first file. The files are next to each other, as you say, but in reverse order.

So I think that optimization by prefetch scenario data will be a big improvement.

Quote from: "Lundholm"
...and what would you do, if a file was shared by several applications?

Good question. The system DLL's are a good example. I haven't thought that far ahead yet, I don't even know the format of the scenario files yet. Perhaps shared files can be placed in a special area before all the other files. Or perhaps they can be placed with the files of the first program that happens to be optimized, and ignored when the optimizer encounters them again in another program. Something like that.
Logged
Lecter
JkDefrag Supporter
***
Posts: 11


View Profile
« Reply #32 on: October 22, 2007, 01:51:19 am »

Can users do that manually? (making the .pf)
Usually I would prefer to respect the FS tree while bringing files with same extensions together.
Logged
Lundholm
JkDefrag Hero
*****
Posts: 208



View Profile
« Reply #33 on: October 22, 2007, 06:40:57 am »

Quote from: "JDPower"
Quote from: "Lundholm"
I think that application files are also moved as part of the XP optimization.
Sorry but they're not.


The issue of prefetch always causes arguments because everyone thinks their way is the right way and there's not enough information from MS to prove either way. So I'm not gonna carry on arguing my point here, what I do works for me and I know it works for me. I don't really care if anyone else disagrees, if what you are doing works for you then keep doing it and be happy :roll:


This a quote from the article referenced in my previous post:
Quote
To minimize seeking even further, every three days or so, during system idle periods, the Task Scheduler organizes a list of files and directories in the order that they are referenced during a boot or application start, and stores the list in a file named \Windows\Prefech\Layout.ini. Figure 1 shows the contents of a prefetch directory, highlighting the layout file. Then it launches the system defragmenter with a command-line option that tells the defragmenter to defragment based on the contents of the file instead of performing a full defrag. The defragmenter finds a contiguous area on each volume large enough to hold all the listed files and directories that reside on that volume and then moves them in their entirety into that area so that they are stored one after the other. Thus, future prefetch operations will even be more efficient because all the data to be read in is now stored physically on the disk in the order it will be read. Since the number of files defragmented for prefetching is usually only in the hundreds, this defragmentation is much faster than full defragmentations.


It looks like you actually care a lot about people disagreeing with you.  Smiley

Cheers
Logged

"O, there has been much throwing about of brains." -- Guildenstern{alt. Gyldenstern[alt. Gyldenstjerne(anc. Gyllenstierna{knight of Lundholm})], knight of Hamlet}.
Lundholm
JkDefrag Hero
*****
Posts: 208



View Profile
« Reply #34 on: October 22, 2007, 07:38:56 am »

Quote from: "jeroen"
Secondly, the "-a 9" optimization will sort the last accessed files to the begin of the disk. If a program uses 2 files when it starts up, then the second file will be placed on disk before the first file. The files are next to each other, as you say, but in reverse order.

So I think that optimization by prefetch scenario data will be a big improvement.


This is a scenario where we can actually benefit from the prefetch function, which will read all the files concurrently, and the disk driver will optimize the read operations.

The best thing JKD can do, is to place the files close to each other, which it already does   Smiley

Thanks again.
Logged

"O, there has been much throwing about of brains." -- Guildenstern{alt. Gyldenstern[alt. Gyldenstjerne(anc. Gyllenstierna{knight of Lundholm})], knight of Hamlet}.
Pages: 1 2 [3]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.5 | SMF © 2006-2008, Simple Machines LLC Valid XHTML 1.0! Valid CSS!