Does anyone else remember...
-
one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? (I'm typing this while waiting for my XP Pro with NTFS workstation to defrag)
Yes, I remember that and I have tried the trial versions of Diskeeper and other programs and found out that they really can't get it totally optimized. I have seen disks that are 70% fragmented and after a complete pass with diskeeper it was 68% fragmented. I emailed them about this. If you are to give a trial version of a product, you want it to perform the task well. Reducing only 2% of the fragmentation of my drive in one pass does not give me any confidence in the product at all. How many passes are needed to get 0% fragmentation or is this imposssible? John
-
one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? (I'm typing this while waiting for my XP Pro with NTFS workstation to defrag)
John Cardinal wrote: one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? Yes, and no, I never defrag! Never enough free space to do it :) leppie::AllocCPArticle(Generic DFA State Machine for .NET);
-
I vaguely remember that claim, which is one reason that I decided to use NTFS on the server I installed for my last employer. As performance steadily dropped in the first year I found out from other sources than Microsoft that NTFS is notorious for fragmentation, worse than FAT, and more than one sysadmin I talked too was surprised that the system still ran at all after a whole year. Naturally, WinNT didn't have a defrag utility, and management wouldn't let me buy a copy of any standard defrag program. That didn't stop their whining about the server's performance, though.
"The Lion shall lie down with the Lamb;
but the Lamb will not get much sleep..."
Lazarus LongI also remember a claim that windows would automatically defragment fs' in the background. Matt Newman
Sonork: 100:11179 "Two things have come out of Berkley, Unix and Acid, we do not belive this to be a coincidence"
Linux sucks twice as fast and 10 times more reliably, and since you have the source, it's your fault. -Ca1v1n Post best viewed with lynx -
Yes, I remember that and I have tried the trial versions of Diskeeper and other programs and found out that they really can't get it totally optimized. I have seen disks that are 70% fragmented and after a complete pass with diskeeper it was 68% fragmented. I emailed them about this. If you are to give a trial version of a product, you want it to perform the task well. Reducing only 2% of the fragmentation of my drive in one pass does not give me any confidence in the product at all. How many passes are needed to get 0% fragmentation or is this imposssible? John
-
I vaguely remember that claim, which is one reason that I decided to use NTFS on the server I installed for my last employer. As performance steadily dropped in the first year I found out from other sources than Microsoft that NTFS is notorious for fragmentation, worse than FAT, and more than one sysadmin I talked too was surprised that the system still ran at all after a whole year. Naturally, WinNT didn't have a defrag utility, and management wouldn't let me buy a copy of any standard defrag program. That didn't stop their whining about the server's performance, though.
"The Lion shall lie down with the Lamb;
but the Lamb will not get much sleep..."
Lazarus LongRoger Wright wrote: NTFS is notorious for fragmentation, worse than FAT My own experience is that it is better than FAT. It may appear worse since the huge partitions of NTFS hold tens of thousands of more files. Still, I recall seeing far more absolute fragmentation on my old 30MB FAT drives than I do today. (I recall running defrag sessions that lasted hours and this was on the fastest machines at the time--on the flip side, the last FAT defraggers MS shipped seemed much better than the crap included with W2K and XP.) One thing I have noticed with NTFS is that while the number of fragmented files on my system is relatively small, they tend to be extremely fragmented.
-
one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? (I'm typing this while waiting for my XP Pro with NTFS workstation to defrag)
John Cardinal wrote: one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? (I'm typing this while waiting for my XP Pro with NTFS workstation to defrag) The allocation algorithm was changed, IIRC, but that claim is too strong on any fs that allows dynamic file growth (have you worked on a mainframe?). When I was on college I wrote a disk defragmenter as a toy project (you need to see how many machines I f***ed up while testing this thing :) ). At that time, I wrote a "disk fragmenter" too, which created and deleted several small files to create some highly fragmented files. This "disk fragmenter" doesn't work anymore, and that says a lot about NTFS allocation algorithm. ORACLE One Real A$#h%le Called Lary Ellison
-
Yes, I remember that and I have tried the trial versions of Diskeeper and other programs and found out that they really can't get it totally optimized. I have seen disks that are 70% fragmented and after a complete pass with diskeeper it was 68% fragmented. I emailed them about this. If you are to give a trial version of a product, you want it to perform the task well. Reducing only 2% of the fragmentation of my drive in one pass does not give me any confidence in the product at all. How many passes are needed to get 0% fragmentation or is this imposssible? John
As long as there are programs accessing the drive other than the defrag utility it is not possible to get 0% fragmentation. You would need a utility that could boot the system and have exclusive access to every file. Defrag tools seem to be a little different these days, they aren't really fixing fragmentation like they used to because Windows uses indexed allocation in the filesystem. If I recall correctly...These file systems have no external fragmentation and internal fragmentation can not be addressed by software. All a defrag tool does now a days is line up disk blocks in a sequential order so that reading the blocks is faster. As always... I could be wrong -Scott
-
As long as there are programs accessing the drive other than the defrag utility it is not possible to get 0% fragmentation. You would need a utility that could boot the system and have exclusive access to every file. Defrag tools seem to be a little different these days, they aren't really fixing fragmentation like they used to because Windows uses indexed allocation in the filesystem. If I recall correctly...These file systems have no external fragmentation and internal fragmentation can not be addressed by software. All a defrag tool does now a days is line up disk blocks in a sequential order so that reading the blocks is faster. As always... I could be wrong -Scott
Scott Lee wrote: You would need a utility that could boot the system and have exclusive access to every file. Indeed, this is what Diskeeper's "Boot-Time Defrag" does. IIRC, it's the only way it can move files out of the MFT on NTFS partitions. - Mike
-
As long as there are programs accessing the drive other than the defrag utility it is not possible to get 0% fragmentation. You would need a utility that could boot the system and have exclusive access to every file. Defrag tools seem to be a little different these days, they aren't really fixing fragmentation like they used to because Windows uses indexed allocation in the filesystem. If I recall correctly...These file systems have no external fragmentation and internal fragmentation can not be addressed by software. All a defrag tool does now a days is line up disk blocks in a sequential order so that reading the blocks is faster. As always... I could be wrong -Scott
mmmm, I have made a very trimmed version of XP that boots from a CD-ROM, Possibly I could defrag from the CD the HDs Will test later. Regardz Colin J Davies
*** WARNING *
This could be addictive
**The minion's version of "Catch :bob: "It's a real shame that people as stupid as you can work out how to use a computer. said by Christian Graus in the Soapbox
-
John Cardinal wrote: one of the promises of the (at the time) new NTFS file system was that it wouldn't ever need to be defragged? (I'm typing this while waiting for my XP Pro with NTFS workstation to defrag) The allocation algorithm was changed, IIRC, but that claim is too strong on any fs that allows dynamic file growth (have you worked on a mainframe?). When I was on college I wrote a disk defragmenter as a toy project (you need to see how many machines I f***ed up while testing this thing :) ). At that time, I wrote a "disk fragmenter" too, which created and deleted several small files to create some highly fragmented files. This "disk fragmenter" doesn't work anymore, and that says a lot about NTFS allocation algorithm. ORACLE One Real A$#h%le Called Lary Ellison
Daniel Turini wrote: I wrote a "disk fragmenter" too, The NU team had a utility called "slow disk" which would fragment a disk. Cute name since it's the opposite of the name of the defragger: Speed Disk :) --Mike-- "I'm working really, really fast at the moment, so a 3 minute outage becomes, due to time dilation, a 5 minute outage." -- Chris Maunder, relativistic system administrator Ericahist | Homepage | RightClick-Encrypt | 1ClickPicGrabber