Re: Compiling for COSS (was Re: squid-2.4stable2 ?)

Henrik Nordstrom wrote:

> Joe Cooper wrote:
>
>
>>Is noatime still appropriate/necessary for a COSS object store? I'm
>>guessing it would be (and I can't test it on my current test machine
>>since my COSS dir is on a normal partition with other normal system stuff).
>
> noatime should not matter much for COSS. It is a problem for the "ufs" family
> of stores because of the amount of inodes being touched by atime, but for
> COSS there only is one inode.

You're right, as always. I was thinking we were still doing lots of
modifications to the file...but it's only being opened once, rather than
every time a new object is inserted, so no constant updating of the
atime. (I'm reading chapter 4 of Stevens APUE which reminded me of this
today.)

>>Interrupts are still going extraordinarily high on write flushes (like
>>12000-16000 showing up on a 'vmstat 1'). Context switches aren't bad
>>though...I would have thought not being kernel level aio would cause
>>extra context switches...but then I'm ignorant on this subject so can't
>>really judge how things ought to be behaving.
>>
>
> Are you using IDE PIO? If so, try DMA.

Sometimes you stun me spotting these things from afar, Henrik. Being a
new machine, which was not originally destined for being a test box
(it's going to find it's way into a colo facility soon to be the company
webserver), it never had DMA enabled. Interrupts are way down with DMA
enabled and CPU usage has dropped a bit too. I'm pushing it to 100
right now, with two drives, and it's looking very good.

Apologies, Adrian. It's like I'm /trying/ to make COSS look worse than
it is, eh?
                                   --
                      Joe Cooper <joe@swelltech.com>
                  Affordable Web Caching Proxy Appliances
                         http://www.swelltech.com
Received on Mon Aug 20 2001 - 23:20:38 MDT