diff options
| author | George Dunlap <george.dunlap@eu.citrix.com> | 2012-10-26 16:42:23 +0100 |
|---|---|---|
| committer | George Dunlap <george.dunlap@eu.citrix.com> | 2012-10-26 16:42:23 +0100 |
| commit | 8159a4585a0b83e4b31e287b64d5c23f61a94cc3 (patch) | |
| tree | a0812177e2a068e492b3c901dc995d9897cf140b /docs | |
| parent | a0fe46bdedff9b7175cdedad1ee29451d4105d35 (diff) | |
| download | xen-8159a4585a0b83e4b31e287b64d5c23f61a94cc3.tar.gz xen-8159a4585a0b83e4b31e287b64d5c23f61a94cc3.tar.bz2 xen-8159a4585a0b83e4b31e287b64d5c23f61a94cc3.zip | |
docs: Document scheduler-related Xen command-line options
Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
xen-unstable changeset: 25968:91e8fd3cf266
Backport-requested-by: Ian Campbell <Ian.Campbell@citrix.com>
Committed-by: Ian Jackson <ian.jackson@eu.citrix.com>
Diffstat (limited to 'docs')
| -rw-r--r-- | docs/misc/xen-command-line.markdown | 22 |
1 files changed, 22 insertions, 0 deletions
diff --git a/docs/misc/xen-command-line.markdown b/docs/misc/xen-command-line.markdown index eff930f7f8..2963d68c2c 100644 --- a/docs/misc/xen-command-line.markdown +++ b/docs/misc/xen-command-line.markdown @@ -701,12 +701,34 @@ Choose the default scheduler. ### sched\_credit\_tslice\_ms > `= <integer>` +Set the timeslice of the credit1 scheduler, in milliseconds. The +default is 30ms. Reasonable values may include 10, 5, or even 1 for +very latency-sensitive workloads. + ### sched\_ratelimit\_us > `= <integer>` +In order to limit the rate of context switching, set the minimum +amount of time that a vcpu can be scheduled for before preempting it, +in microseconds. The default is 1000us (1ms). Setting this to 0 +disables it altogether. + ### sched\_smt\_power\_savings > `= <boolean>` +Normally Xen will try to maximize performance and cache utilization by +spreading out vcpus across as many different divisions as possible +(i.e, numa nodes, sockets, cores threads, &c). This often maximizes +throughput, but also maximizes energy usage, since it reduces the +depth to which a processor can sleep. + +This option inverts the logic, so that the scheduler in effect tries +to keep the vcpus on the smallest amount of silicon possible; i.e., +first fill up sibling threads, then sibling cores, then sibling +sockets, &c. This will reduce performance somewhat, particularly on +systems with hyperthreading enabled, but should reduce power by +enabling more sockets and cores to go into deeper sleep states. + ### serial\_tx\_buffer > `= <size>` |