Re: Optimal filesystem block sizes for cache

From: B. Richardson <rabtter@dont-contact.us>
Date: Tue, 21 Jul 1998 21:33:54 -0400 (EDT)

On Tue, 21 Jul 1998, Alex Rousskov wrote:

> On Tue, 21 Jul 1998, B. Richardson wrote:
>
> > I made the block sizes 4K with out doing a dissertation on VM disk
> > cacheing or the internals of squid, but buy simply looking at the
> > read histogram via the cachemgr and noticing that %80 of the reads
> > are 4K or less.
>
> Right. I am curious if you have noticed any change in Squid performance after
> the block size was altered? If there was a positive trend, maybe others (with
> similar environments) should follow your example..
>
> Alex.
>
>

A measurable performance increase I'm not sure of, but under heavy
load the box feels more responsive for interactive logins. Unfortunately
the powers that be are convinced the box does "better" and the
timed involved to collect informative data of 4096 vs 8192 (or other)
blocksize is unnecessary :-( (they just don't understand that I
built the box for me to play with :-)).

Barrett
Received on Tue Jul 21 1998 - 18:33:46 MDT

This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:41:13 MST