linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Rik van Riel <H.H.vanRiel@fys.ruu.nl>
To: Linus Torvalds <torvalds@transmeta.com>
Cc: "Stephen C. Tweedie" <sct@dcs.ed.ac.uk>, linux-mm <linux-mm@kvack.org>
Subject: new allocation algorithm
Date: Fri, 27 Mar 1998 10:03:37 +0100 (MET)	[thread overview]
Message-ID: <Pine.LNX.3.91.980327095733.3532A-100000@mirkwood.dummy.home> (raw)

Hi Linus and Stephen,

I just came up with the idea of using an ext2 like algorithm
for memory allocation, in which we:
- group memory in 128 PAGE groups
- have one unsigned char counter per group, counting the number
  of used pages
- allocate with the goal to fill the busiest group
- penalize pages (page->age) when the group is relatively empty
- thereby shift memory usage from empty groups to full groups, making
  larger allocation in empty groups easier
- PLUS we get rid of the 'randomness' in the current system, where we
  sometimes swap out over half of total memory in one sweep while
  still not reaching our goals

Of course, such an algorithm would be relatively expensive...
But could it be worth it?

Trading off the tradeoffs,

Rik.
+-------------------------------------------+--------------------------+
| Linux: - LinuxHQ MM-patches page          | Scouting       webmaster |
|        - kswapd ask-him & complain-to guy | Vries    cubscout leader |
|     http://www.fys.ruu.nl/~riel/          | <H.H.vanRiel@fys.ruu.nl> |
+-------------------------------------------+--------------------------+

             reply	other threads:[~1998-03-27 10:03 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
1998-03-27  9:03 Rik van Riel [this message]
1998-03-27 17:30 ` Linus Torvalds
1998-03-27 18:33   ` Benjamin C.R. LaHaise
1998-03-30 15:07   ` Rik van Riel
1998-04-01 20:28   ` Stephen C. Tweedie

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=Pine.LNX.3.91.980327095733.3532A-100000@mirkwood.dummy.home \
    --to=h.h.vanriel@fys.ruu.nl \
    --cc=linux-mm@kvack.org \
    --cc=sct@dcs.ed.ac.uk \
    --cc=torvalds@transmeta.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox