linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Jonathan Lahr <lahr@sequent.com>
To: Anton Blanchard <anton@linuxcare.com.au>
Cc: Jonathan Lahr <lahr@sequent.com>, linux-kernel@vger.kernel.org
Subject: Re: kernel lock contention and scalability
Date: Tue, 6 Mar 2001 14:45:52 -0800	[thread overview]
Message-ID: <20010306144552.G6451@w-lahr.des.sequent.com> (raw)
In-Reply-To: <20010215104656.A6856@w-lahr.des.sequent.com> <20010305113807.A3917@linuxcare.com>
In-Reply-To: <20010305113807.A3917@linuxcare.com>; from anton@linuxcare.com.au on Mon, Mar 05, 2001 at 11:38:08AM +1100


> Tridge and I tried out the postgresql benchmark you used here and this
> contention is due to a bug in postgres. From a quick strace, we found
> the threads do a load of select(0, NULL, NULL, NULL, {0,0}). Basically all
> threads are pounding on schedule().
...
> Our guess is that the app has some form of userspace synchronisation
> (semaphores/spinlocks). I'd argue that the app needs to be fixed not the
> kernel, or a more valid test case is put forwards. :)
...
> PS: I just looked at the postgresql source and the spinlocks (s_lock() etc)
> are in a tight loop doing select(0, NULL, NULL, NULL, {0,0}). 

Anton,

Thanks for looking into postgresql/pgbench related locking.  Yes, 
apparently postgresql uses a synchronization scheme that uses select()
to effect delays for backing off while attempting to acquire a lock.
However, it seems to me that runqueue lock contention was not entirely due 
to postgresql code, since it was largely alleviated by the multiqueue 
scheduler patch.

In using postgresql/pgbench to measure lock contention, I was attempting
to apply a typical server workload to measure scalability using only open 
software.  My goal is to load and measure the kernel for server performance, 
so I need to ensure that the software I use represents likely real world 
server configurations.  I did not use mysql, because it cannot perform 
transactions which I considered important.  Any pointers to other open 
database software or benchmarks that might be suitable for this effort 
would be appreciated.

Jonathan


  reply	other threads:[~2001-03-06 22:51 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2001-02-15 18:46 kernel lock contention and scalability Jonathan Lahr
2001-02-25  9:52 ` Manfred Spraul
2001-03-05 18:41   ` Jonathan Lahr
2001-03-05  0:38 ` Anton Blanchard
2001-03-06 22:45   ` Jonathan Lahr [this message]
2001-03-06 23:39     ` Matthew Kirkwood
2001-03-07  0:28       ` Tim Wright
2001-03-07  3:12         ` Jeff Dike
2001-03-07 22:13           ` Tim Wright
2001-03-08 23:26             ` Jeff Dike
2001-03-11  6:50       ` Anton Blanchard
2001-03-11  6:26     ` Anton Blanchard
     [not found] <98454d$19p9h$1@fido.engr.sgi.com>
2001-03-07  2:55 ` Rajagopal Ananthanarayanan
2001-03-07  5:48   ` Jeff Dike

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20010306144552.G6451@w-lahr.des.sequent.com \
    --to=lahr@sequent.com \
    --cc=anton@linuxcare.com.au \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).