>>> Arjan de Vet wrote
> I'm working on an updated and improved version of my Squid timing patch.
> What I see on a very busy cache (100,000 conns/hour) is that open(2) is
> *the* bottleneck (50-60% of the time :-() and we should have something
> like an opend or seperate processes which do all disk I/O so Squid doesn't
> stop during blocking disk operations.
Stew Forster's asyncio changes (which are in 1.2, AFAIK) make all of the
open() (and other) system calls be handled asynchronously. It's much neater
to do this than to have a seperate process. (well, neater in my eyes, anyway :)
Anthony
Received on Tue Jul 29 2003 - 13:15:44 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:11:28 MST