> We're not sure if it is something we've done or just increased usage, but
> bkbits.net is getting hammered lately. We see load averages in ~15-20
> range pretty regularly. It's got some nasty characteristics from the
> point of view of server/VM system, tons of data and not really any good
> working sets.
There are a few things that could help here:
1) Andrew Morton's io scheduler patch (read-latency2), which is
in the -rmap kernel
2) reducing the number of outstanding IO commands in the 3ware
controller/driver, so akpm's io scheduler gets a chance to do
its work
3) better readahead, if possible...
regards,
Rik
-- Bravely reimplemented by the knights who say "NIH".http://www.surriel.com/ http://distro.conectiva.com/
Spamtraps of the month: september@surriel.com trac@trac.org
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/