Also seeing 2 x quad-core system slower that 2 x dual core

Pete French petefrench at ticketswitch.com
Thu Nov 29 16:08:14 PST 2007


> Can you provide more details on this task?  It seems like something that
> could easily be reproduced in a lab environment and serve as a regression
> test and baseline for future improvements.  Is the server doing any other
> work while doing the rm, or is this it?  What kind of directory layout
> are we looking at?  How many files, directories, users ...?  What type of
> filesystem, options, size?

It's a simple 'rm -rf /usr/local/etc/www/live'. The system is not running
apache when I do that, though it is still running mysql. Nothing else
going on. Filesystems are all UFS2, with default options as created
by sysinstall - except I have softupdates on. Disc is partitioned
as 1024M for /var/log, 64853M for / (with soft updates) and 4096M for swap
at the end (it never swaps). The directory is 1 gig of web files, 165,000
of them or so.

> BTW, what's your nsswitch.conf like?  If you don't need nis, removing it=20
> can mean a considerable speed improvement - or try nscd(8).

It's the default as installed. havent even even looked at that file before,
but I dont need NIS, no.

Thing is that GENERIC as installed out of the box should not take two minutes
to delete a gig of files off a 15k RPM SAS drive! especially not
when identical hardware with half the number of processor cores only takes
eleven seconds to do the same job. Something is wrong somewhere if doubling
your CPU's results in a factor of 12 slowdown surely?

-pete.


More information about the freebsd-stable mailing list