All of lore.kernel.org
 help / color / mirror / Atom feed
From: Con Kolivas <kernel@kolivas.org>
To: Peter Williams <pwil3058@bigpond.net.au>
Cc: spaminos-ker@yahoo.com,
	William Lee Irwin III <wli@holomorphy.com>,
	linux-kernel@vger.kernel.org
Subject: Re: Scheduler fairness problem on 2.6 series (Attn: Nick Piggin and others)
Date: Wed, 11 Aug 2004 13:31:16 +1000	[thread overview]
Message-ID: <cone.1092195076.205601.25569.502@pc.kolivas.org> (raw)
In-Reply-To: 41199129.9080809@bigpond.net.au

Peter Williams writes:

> Peter Williams wrote:
>> Peter Williams wrote:
>> 
>>> William Lee Irwin III wrote:
>>>
>>>> On Tue, Aug 10, 2004 at 07:21:43PM -0700, spaminos-ker@yahoo.com wrote:
>>>>
>>>>> I am not very familiar with all the parameters, so I just kept the 
>>>>> defaults
>>>>> Anything else I could try?
>>>>> Nicolas
>>>>
>>>>
>>>>
>>>>
>>>> No. It appeared that the SPA bits had sufficient fairness in them to
>>>> pass this test but apparently not quite enough.
>>>>
>>>
>>> The interactive bonus may interfere with fairness (the throughput 
>>> bonus should actually help it for tasks with equal nice) so you could 
>>> try setting max_ia_bonus to zero (and possibly increasing 
>>> max_tpt_bonus). With "eb" mode this should still give good interactive 
>>> response but expect interactive response to suffer a little in "pb" 
>>> mode however renicing the X server to a negative value should help.
>> 
>> 
>> I should also have mentioned that fiddling with the promotion interval 
>> may help.
> 
> Having reread your original e-mail I think that this problem is probably 
>   being caused by the interactive bonus mechanism classifying the httpd 
> server threads as "interactive" threads and giving them a bonus.  But 
> for some reason the daemon is not identified as "interactive" meaning 
> that it gets given a lower priority.  In this situation if there's a 
> large number of httpd threads (even with promotion) it could take quite 
> a while for the daemon to get a look in.  Without promotion total 
> starvation is even a possibility.
> 
> Peter
> PS For both "eb" and "pb" modes, max_io_bonus should be set to zero on 
> servers (where interactive responsiveness isn't an issue).
> PPS For "sc" mode, try setting "interactive" to zero and "compute" to 1.

No, compute should not be set to 1 for a server. It is reserved only for 
computational nodes, not regular servers. "Compute"  will increase latency 
which is undersirable.

Cheers,
Con


  reply	other threads:[~2004-08-11  3:31 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20040811010116.GL11200@holomorphy.com>
2004-08-11  2:21 ` Scheduler fairness problem on 2.6 series (Attn: Nick Piggin and others) spaminos-ker
2004-08-11  2:23   ` William Lee Irwin III
2004-08-11  2:45     ` Peter Williams
2004-08-11  2:47       ` Peter Williams
2004-08-11  3:23         ` Peter Williams
2004-08-11  3:31           ` Con Kolivas [this message]
2004-08-11  3:46             ` Peter Williams
2004-08-11  3:44           ` Peter Williams
2004-08-13  0:13             ` spaminos-ker
2004-08-13  1:44               ` Peter Williams
2004-08-11  3:09   ` Con Kolivas
2004-08-11 10:24     ` Prakash K. Cheemplavam
2004-08-11 11:26       ` Scheduler fairness problem on 2.6 series Con Kolivas
2004-08-11 12:05         ` Prakash K. Cheemplavam
2004-08-11 19:22           ` Prakash K. Cheemplavam
2004-08-11 23:42             ` Con Kolivas
2004-08-12  8:08               ` Prakash K. Cheemplavam
2004-08-12 18:18               ` Bill Davidsen
2004-08-12  2:04     ` Scheduler fairness problem on 2.6 series (Attn: Nick Piggin and others) spaminos-ker
2004-08-12  2:24     ` spaminos-ker
2004-08-12  2:53       ` Con Kolivas
     [not found] <411D50AE.5020005@bigpond.net.au>
2004-08-17 23:19 ` spaminos-ker
2004-08-18  0:12   ` Peter Williams
2004-08-24 21:11     ` spaminos-ker
2004-08-24 23:04       ` Peter Williams
2004-08-24 23:22         ` Lee Revell
2004-08-26  2:30         ` spaminos-ker
2004-08-26  2:42           ` Peter Williams
2004-08-26  8:39             ` Peter Williams
2004-08-28  1:59               ` spaminos-ker
2004-08-29  0:21                 ` Peter Williams
2004-08-29  0:25                   ` Lee Revell
2004-08-29  0:45                     ` Lee Revell
2004-08-29  2:03                       ` Peter Williams
2004-08-29  2:28                         ` spaminos-ker
2004-08-29  4:53                           ` Peter Williams
2004-08-29  1:19                     ` spaminos-ker
2004-08-29  1:22                       ` Lee Revell
2004-08-29  1:31                         ` Peter Williams
2004-09-13 20:09                           ` spaminos-ker
2004-08-29  2:20                       ` Lee Revell
     [not found] <20040811093945.GA10667@elte.hu>
2004-08-17 23:08 ` spaminos-ker
2004-08-07 21:53 spaminos-ker

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=cone.1092195076.205601.25569.502@pc.kolivas.org \
    --to=kernel@kolivas.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pwil3058@bigpond.net.au \
    --cc=spaminos-ker@yahoo.com \
    --cc=wli@holomorphy.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.