Defrag for WinNT 4.0

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

d.086@hotmail.com - 28 Mar 2007 22:40 GMT

Hi,

I'm maintaining a data system that is part of a scientific instrument.
The instrument software is set up to run under MS Windows 4.0 with
Service Pack 3. It is not wise to change it. I've set up the drive for
a FAT partition. I want to leave it as a FAT partition. My question:
Is there a (free or freeware) defrag tool that I can download (where?)
that will work with this configuration? I've tried defrag under MSDOS
6.22 and that was a disaster, all the long file names were lost and I
had to reinstall all the sofware. I tried the defrag tool for MS
Windows 98 that works with FAT partitions and preserves long file
names but it wouldn't start under WinNT 4.0. Thanks for the help.

d.

Reply to this Message

John John - 29 Mar 2007 00:24 GMT

Diskeeper Lite version 7.0.418 has served me well. You can download it
here: http://www.majorgeeks.com/Diskeeper_Lite_d1207.html

One thing that I don't like about it is that after you run it and then
exit or close it, it keeps a running process going and it can't be
killed unless you reboot. On critical machines I learned to live with
that by simply running it near to when shutdowns are scheduled.

John

> Hi,
>
[quoted text clipped - 10 lines]
>
> d.

Reply to this Message

Dean Dark - 29 Mar 2007 00:41 GMT

>I'm maintaining a data system that is part of a scientific instrument.
>The instrument software is set up to run under MS Windows 4.0 with
[quoted text clipped - 6 lines]
>Windows 98 that works with FAT partitions and preserves long file
>names but it wouldn't start under WinNT 4.0. Thanks for the help.

What is driving your belief that you need to defrag. the drive on the
system in the first place? Do you have evidence that file
fragmentation is causing performance problems?

Just asking...

Signature

Dan.

Reply to this Message

d.086@hotmail.com - 29 Mar 2007 01:23 GMT

> >I'm maintaining a data system that is part of a scientific instrument.
> >The instrument software is set up to run under MS Windows 4.0 with
[quoted text clipped - 14 lines]
> --
> Dan.

Just good practice, I believe. When I did a defrag using MSDOS 6.22
the defrag program showed an 8% level of fragmentation. MSDOS 6.22
made the system unusable because it did not preserve long file names.
However, the system did boot into WinNT 4.0 and I got the impression
the boot time was a wee bit quicker. But it's possible that it booted
a wee bit quicker since some of the system to be loaded at boot time
did not load. In various instances in the past, I've found that
defragging the file system can improve performance and file system
reliability. If a defrag can be done after installing OS and
application software, it is a good idea to do it, I think. Also,
defrag (like scandisk error checking), tests the hard disk hardware
more extensively. That is important for a scientific instrument. One
doesn't want to be performing a critical scientific experiment, only
to discover that the hard drive is flaky. Fragmentation of the free
space area can cause new files to be fragmented. The data files
generated by the instrument sofware can be large and I think it would
be a good idea to help prevent these files from being fragmented. I
may be wrong, but that's what I believe.

Reply to this Message

Dean Dark - 29 Mar 2007 12:33 GMT

>> What is driving your belief that you need to defrag. the drive on the
>> system in the first place? Do you have evidence that file
>> fragmentation is causing performance problems?

>Just good practice, I believe. When I did a defrag using MSDOS 6.22
>the defrag program showed an 8% level of fragmentation. MSDOS 6.22
[quoted text clipped - 14 lines]
>be a good idea to help prevent these files from being fragmented. I
>may be wrong, but that's what I believe.

My experience with validated instrumentation and data acquisition
systems in pharmaceutical and biotech manufacturing (much of it using
so called "legacy" systems running NT4 and earlier more arcane OSs) is
that disk defragmentation is a "comfort food" that has little if any
nutritional value. On the other hand, if you can actually *prove*
that disk fragmentation is *crippling* your system, then you should
tackle the cause of it, and not just try to fix the symptoms.

If you have the luxury of not needing your systems to run 24/7 and you
are worried about disk errors, then you should probably run checkdisk
once a month or so.

Signature

Dan.

Reply to this Message

John John - 29 Mar 2007 14:45 GMT

>>>What is driving your belief that you need to defrag. the drive on the
>>>system in the first place? Do you have evidence that file
[quoted text clipped - 26 lines]
> that disk fragmentation is *crippling* your system, then you should
> tackle the cause of it, and not just try to fix the symptoms.

I agree. I have DKLite on my NT4 boxes but I can't remember when I last
ran it. The only time I might run it is if I move or delete lots of
files or uninstall big programs. On my mature production boxes the
later is rather infrequent.

John

Reply to this Message

Joshua Bolton - 08 May 2007 19:14 GMT

clearly John and Dean your systems are using databases and not
writing/deleting files. If they were and you let your ntfs systems become
overly fragmented you can render your systems inoperable.

It is ALWAYS recommended as part of a regular maintenance routine to
defragment your drives.

Reply to this Message