CPU load due to having drives on a server compressed.

  • Thread starter Thread starter aengus
  • Start date Start date
A

aengus

Guest
Hi,

A drive on a Windows 2003 server is compressed and a user is reading
and writing files on that drive over the LAN from his desktop
computer. Which computer is doing the processing associated with
compressing and decompressing the data? Is it the server, or the
user's desktop machine?

Thanks,

Aengus.
 
Re: CPU load due to having drives on a server compressed.

user's

--

Regards,

Dave Patrick ....Please no email replies - reply in newsgroup.
Microsoft Certified Professional
Microsoft MVP [Windows]
http://www.microsoft.com/protect

"aengus" wrote:
> Hi,
>
> A drive on a Windows 2003 server is compressed and a user is reading
> and writing files on that drive over the LAN from his desktop
> computer. Which computer is doing the processing associated with
> compressing and decompressing the data? Is it the server, or the
> user's desktop machine?
>
> Thanks,
>
> Aengus.
>
 
Re: CPU load due to having drives on a server compressed.

Hi aengus,

The server is doing all the work. The user may notice a speed decrease while
opening or saving files on a compressed drive but this is due to the server
and network bandwidth.
http://support.microsoft.com/kb/251186

Coraleigh Miller


"aengus" <aengus@gmail.com> wrote in message
news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
> Hi,
>
> A drive on a Windows 2003 server is compressed and a user is reading
> and writing files on that drive over the LAN from his desktop
> computer. Which computer is doing the processing associated with
> compressing and decompressing the data? Is it the server, or the
> user's desktop machine?
>
> Thanks,
>
> Aengus.
>
 
Re: CPU load due to having drives on a server compressed.

Hi again aengus,

I defer to Dave Patrick on this one. :-)
I just ran perfmon on my server and client and setup a quick test to see
what happened and sure enough my client's CPU and IO spiked while the file
server stayed level.

Thanks Dave!

Coraleigh


"aengus" <aengus@gmail.com> wrote in message
news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
> Hi,
>
> A drive on a Windows 2003 server is compressed and a user is reading
> and writing files on that drive over the LAN from his desktop
> computer. Which computer is doing the processing associated with
> compressing and decompressing the data? Is it the server, or the
> user's desktop machine?
>
> Thanks,
>
> Aengus.
>
 
Re: CPU load due to having drives on a server compressed.

There's two conflicting answers! The KB article seems to indicate that
the server is doing all the work.

Thanks,

Aengus.


On Aug 20, 1:15 pm, "Coraleigh Miller" <CoraleighMil...@yahoo.com>
wrote:
> Hi aengus,
>
> The server is doing all the work. The user may notice a speed decrease while
> opening or saving files on a compressed drive but this is due to the server
> and network bandwidth.http://support.microsoft.com/kb/251186
>
> Coraleigh Miller
>
> "aengus" <aen...@gmail.com> wrote in message
>
> news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
>
> > Hi,
>
> > A drive on a Windows 2003 server is compressed and a user is reading
> > and writing files on that drive over the LAN from his desktop
> > computer. Which computer is doing the processing associated with
> > compressing and decompressing the data? Is it the server, or the
> > user's desktop machine?
>
> > Thanks,
>
> > Aengus
 
Re: CPU load due to having drives on a server compressed.

Ok, fair enough...

But the KB article _does_ suggest differently doesn't it? In the case
of decompression, the article says "Compressed files are also expanded
before copying over the network", how could the user's computer do the
work in this case?

Cheers,

Aengus.



On Aug 20, 1:38 pm, "Coraleigh Miller" <CoraleighMil...@yahoo.com>
wrote:
> Hi again aengus,
>
> I defer to Dave Patrick on this one. :-)
> I just ran perfmon on my server and client and setup a quick test to see
> what happened and sure enough my client's CPU and IO spiked while the file
> server stayed level.
>
> Thanks Dave!
>
> Coraleigh
>
> "aengus" <aen...@gmail.com> wrote in message
>
> news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
>
> > Hi,
>
> > A drive on a Windows 2003 server is compressed and a user is reading
> > and writing files on that drive over the LAN from his desktop
> > computer. Which computer is doing the processing associated with
> > compressing and decompressing the data? Is it the server, or the
> > user's desktop machine?
>
> > Thanks,
>
> > Aengus.
 
Re: CPU load due to having drives on a server compressed.

It does indeed! Out of my respect for Dave Patrick however I decided to
perform my own little test as well as gave a closer inspection to the
article and it doesnt actually cite our situation exactly. Try running a
perfmon test on your server and client after hours to see what you come up
with.

Coraleigh


"aengus" <aengus@gmail.com> wrote in message
news:1187582144.752280.291190@l22g2000prc.googlegroups.com...
> Ok, fair enough...
>
> But the KB article _does_ suggest differently doesn't it? In the case
> of decompression, the article says "Compressed files are also expanded
> before copying over the network", how could the user's computer do the
> work in this case?
>
> Cheers,
>
> Aengus.
>
>
>
> On Aug 20, 1:38 pm, "Coraleigh Miller" <CoraleighMil...@yahoo.com>
> wrote:
>> Hi again aengus,
>>
>> I defer to Dave Patrick on this one. :-)
>> I just ran perfmon on my server and client and setup a quick test to see
>> what happened and sure enough my client's CPU and IO spiked while the
>> file
>> server stayed level.
>>
>> Thanks Dave!
>>
>> Coraleigh
>>
>> "aengus" <aen...@gmail.com> wrote in message
>>
>> news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
>>
>> > Hi,
>>
>> > A drive on a Windows 2003 server is compressed and a user is reading
>> > and writing files on that drive over the LAN from his desktop
>> > computer. Which computer is doing the processing associated with
>> > compressing and decompressing the data? Is it the server, or the
>> > user's desktop machine?
>>
>> > Thanks,
>>
>> > Aengus.
>
>
 
Re: CPU load due to having drives on a server compressed.

Thank you for the compliment. I also read your article with interest. I
don't see how/why the server would handle the request for
compression/decompression. I'll be interested to see aengus's test result as
well.

--

Regards,

Dave Patrick ....Please no email replies - reply in newsgroup.
Microsoft Certified Professional
Microsoft MVP [Windows]
http://www.microsoft.com/protect

"Coraleigh Miller" wrote:
> It does indeed! Out of my respect for Dave Patrick however I decided to
> perform my own little test as well as gave a closer inspection to the
> article and it doesnt actually cite our situation exactly. Try running a
> perfmon test on your server and client after hours to see what you come up
> with.
>
> Coraleigh
 
Re: CPU load due to having drives on a server compressed.

Now that I know what MVP stands for, it seems Dave Patrick is one to
be trusted on this issue alright!

I'll run a perfmon at some point just to have a look, but I consider
the question answered now.

Thanks again,

Aengus.


On Aug 20, 2:08 pm, "Coraleigh Miller" <CoraleighMil...@yahoo.com>
wrote:
> It does indeed! Out of my respect for Dave Patrick however I decided to
> perform my own little test as well as gave a closer inspection to the
> article and it doesnt actually cite our situation exactly. Try running a
> perfmon test on your server and client after hours to see what you come up
> with.
>
> Coraleigh
>
> "aengus" <aen...@gmail.com> wrote in message
>
> news:1187582144.752280.291190@l22g2000prc.googlegroups.com...
>
> > Ok, fair enough...
>
> > But the KB article _does_ suggest differently doesn't it? In the case
> > of decompression, the article says "Compressed files are also expanded
> > before copying over the network", how could the user's computer do the
> > work in this case?
>
> > Cheers,
>
> > Aengus.
>
> > On Aug 20, 1:38 pm, "Coraleigh Miller" <CoraleighMil...@yahoo.com>
> > wrote:
> >> Hi again aengus,
>
> >> I defer to Dave Patrick on this one. :-)
> >> I just ran perfmon on my server and client and setup a quick test to see
> >> what happened and sure enough my client's CPU and IO spiked while the
> >> file
> >> server stayed level.
>
> >> Thanks Dave!
>
> >> Coraleigh
>
> >> "aengus" <aen...@gmail.com> wrote in message
>
> >>news:1187578481.953393.152810@e9g2000prf.googlegroups.com...
>
> >> > Hi,
>
> >> > A drive on a Windows 2003 server is compressed and a user is reading
> >> > and writing files on that drive over the LAN from his desktop
> >> > computer. Which computer is doing the processing associated with
> >> > compressing and decompressing the data? Is it the server, or the
> >> > user's desktop machine?
>
> >> > Thanks,
>
> >> > Aengus.
 
Re: CPU load due to having drives on a server compressed.

I did a quick test.

During the copying of files from the server to my local machine, the
only noticeable increase in CPU usage was on the local machine. When
copying files to the server, there was a similar increase in local
machine CPU usage, but this seemed to be accompanied by a small
increase in server CPU usage.

Aengus.


On Aug 20, 2:19 pm, "Dave Patrick" <DSPatr...@nospam.gmail.com> wrote:
> Thank you for the compliment. I also read your article with interest. I
> don't see how/why the server would handle the request for
> compression/decompression. I'll be interested to see aengus's test result as
> well.
>
> --
>
> Regards,
>
> Dave Patrick ....Please no email replies - reply in newsgroup.
> Microsoft Certified Professional
> Microsoft MVP [Windows]http://www.microsoft.com/protect
>
> "Coraleigh Miller" wrote:
> > It does indeed! Out of my respect for Dave Patrick however I decided to
> > perform my own little test as well as gave a closer inspection to the
> > article and it doesnt actually cite our situation exactly. Try running a
> > perfmon test on your server and client after hours to see what you come up
> > with.
>
> > Coraleigh
 
Back
Top