From: George Dunlap <george.dunlap@eu.citrix.com>
To: xen-devel@lists.xensource.com
Cc: george.dunlap@eu.citrix.com
Subject: [PATCH 00 of 16] credit2: Scale to multiple sockets
Date: Thu, 23 Dec 2010 12:38:32 +0000 [thread overview]
Message-ID: <patchbomb.1293107912@localhost6.localdomain6> (raw)
This patch series allows credit2 to scale reasonably past one socket.
The to key things it introduces are:
* Code to detect cpu topology, and create one runqueue per socket
* A first-cut at a load-average based load balancer
Quick descriptions below:
01: Make some debug messages quieter.
02-03: Clean up some of the per-cpu runqueue lock pointer code. This
is the only part of the patch that touches code outside the credit2
scheduler.
04-07: Infrastructure for having multiple sockets, including hooks for
load calculation
08: Detect socket layout, make one runqueue per socket
09-13: Infrastructure for load-balancing, including load averages for
runqueues and sockets
14: A load-average-based balancer
15: Refinement on when to balance
16: Debug key output
next reply other threads:[~2010-12-23 12:38 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2010-12-23 12:38 George Dunlap [this message]
2010-12-23 12:38 ` [PATCH 01 of 16] credit2: Quieten some debug messages George Dunlap
2010-12-23 12:38 ` [PATCH 02 of 16] scheduler: Update vcpu_schedule_lock to check for changed lock pointer as well George Dunlap
2010-12-23 12:38 ` [PATCH 03 of 16] scheduler: Introduce pcpu_schedule_lock George Dunlap
2010-12-23 12:38 ` [PATCH 04 of 16] credit2: Refactor runqueue initialization George Dunlap
2010-12-23 12:38 ` [PATCH 05 of 16] credit2: Handle runqueue changes George Dunlap
2010-12-23 12:38 ` [PATCH 06 of 16] credit2: Calculate instantaneous runqueue load George Dunlap
2010-12-23 12:38 ` [PATCH 07 of 16] credit2: Simple cpu picker based on instantaneous load George Dunlap
2010-12-23 12:38 ` [PATCH 08 of 16] credit2: Detect socket layout and assign one runqueue per socket George Dunlap
2010-12-23 12:38 ` [PATCH 09 of 16] credit2: Calculate load average George Dunlap
2010-12-23 12:38 ` [PATCH 10 of 16] credit2: Track average load contributed by a vcpu George Dunlap
2010-12-23 12:38 ` [PATCH 11 of 16] credit2: Track expected load George Dunlap
2010-12-23 12:38 ` [PATCH 12 of 16] credit2: Migrate request infrastructure George Dunlap
2010-12-23 12:38 ` [PATCH 13 of 16] credit2: Use loadavg to pick cpus, instead of instantaneous load George Dunlap
2010-12-23 12:38 ` [PATCH 14 of 16] credit2: Introduce a loadavg-based load balancer George Dunlap
2010-12-23 12:38 ` [PATCH 15 of 16] credit2: Different unbalance tolerance for underloaded and overloaded queues George Dunlap
2010-12-23 12:38 ` [PATCH 16 of 16] credit2: On debug keypress print load average as a fraction George Dunlap
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=patchbomb.1293107912@localhost6.localdomain6 \
--to=george.dunlap@eu.citrix.com \
--cc=xen-devel@lists.xensource.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.