[H-GEN] Poor performance of squid-2.5 on Redhat 7.2 compared to Solaris 8 on x86

[ben.carlyle at invensys.com]

[ Humbug *General* list - semi-serious discussions about Humbug and  ]
[ Unix-related topics.  Please observe the list's charter.           ]
[ Worthwhile understanding: http://www.humbug.org.au/netiquette.html ]

G'day.

During my short stint at the TAB of Queensland we had a curious issue with 
bet processing machines. We had started bringing in new pentiums to 
replace 486 machines that had done the job before. I didn't see the issue 
through to completion, but I was able to prove that it was related to the 
performance of IO through the comms card used. The card was an ISA card, 
and although the new machines supported ISA it was done through some 
quirky PCI-ISA mapping on the board. It appeared that the drivers or 
hardware that handled this was what reduced the processing power of the 
pentium to about 1/10th the performance of the little 486s with 
althogether inferiour specs.

Machine architectures can make big differences :)

Benjamin.






"Michael Anthon" <michael at anthon.net>
Sent by: Majordomo <majordom at caliburn.humbug.org.au>
20/02/02 22:09
Please respond to general

 
        To:     <general at lists.humbug.org.au>
        cc: 
        Subject:        Re: [H-GEN] Poor performance of squid-2.5 on Redhat 7.2 compared to 
Solaris 8 on x86



I'm guessing here and maybe you have already tested this, but are you
certain it is squid that is the problem?  If you have a web server running
on the machine with squid, try access a file from there via squid and see
how it goes.  Maybe it's the network card or it's settings or some other
network related issue.

Cheers
Michael

<snippity snip>
> Anyway we have a url we can access that indicates download speed and for
> some reason the solis box ALWAYS leaves the Redhat box for
dead....1500Kbps
> vs 200-300Kbps
>
<snip>



--
* This is list (humbug) general handled by majordomo at lists.humbug.org.au .
* Postings to this list are only accepted from subscribed addresses of
* lists 'general' or 'general-post'.