Skip to contentSkip to conversation

cfq-iosched: improve latency for no-idle queues

RFC PATCH In review Corrado Zoccolo 2 participants 2 messages Oct 19, 2009

lore
Share & tools
lore
#0 lore raw

cover cfq-iosched: improve latency for no-idle queues

This patch series is intended to improve I/O latency, addressing an often neglected, important subset of workloads: the ones for which cfq currently prefers not to do any idling.

Those are the ones that would benefit most from having low latency, in fact they are any of:
* processes with large think times (e.g. interactive ones like file managers)
* seeky (e.g. programs faulting in their code at startup)
* or marked as no-idle from upper levels.

The patch series addresses this by:
* reducing queues' timeslice when many queues have pending I/O
* separating queues with different priorities and different characteristics in different service trees, each with an allocated time slice
* enable idling when switching between service trees, even for queues that would not have idling enabled otherwise.

This provides various benefits:
* service tree insertion code is simplified, since it doesn't need to cope with priorities any more.
* high priority no_idle queues are no longer penalized when competing with lower priority, idling queues
* seeky and no_idle queues have their fair share of disk time, without penalizing NCQ drives' performances, since they can all dispatch together, filling up the available NCQ slots.

On a non-NCQ capable drive, a workload of 4 random readers competing with sequential writer, the maximum latency experienced by readers decreased from > 500ms to about 160ms.

#1 lore raw
Quoted 26 lines · Corrado Zoccolo

This patch series is intended to improve I/O latency, addressing an often neglected, important subset of workloads: the ones for which cfq currently prefers not to do any idling.

Those are the ones that would benefit most from having low latency, in fact they are any of:
* processes with large think times (e.g. interactive ones like file managers)
* seeky (e.g. programs faulting in their code at startup)
* or marked as no-idle from upper levels.

The patch series addresses this by:
* reducing queues' timeslice when many queues have pending I/O
* separating queues with different priorities and different characteristics in different service trees, each with an allocated time slice
* enable idling when switching between service trees, even for queues that would not have idling enabled otherwise.

This provides various benefits:
* service tree insertion code is simplified, since it doesn't need to cope with priorities any more.
* high priority no_idle queues are no longer penalized when competing with lower priority, idling queues
* seeky and no_idle queues have their fair share of disk time, without penalizing NCQ drives' performances, since they can all dispatch together, filling up the available NCQ slots.

Quoted 3 lines

On a non-NCQ capable drive, a workload of 4 random readers competing with sequential writer, the maximum latency experienced by readers decreased from > 500ms to about 160ms.

Thanks, interesting series. I'll look over the patches as time permits and try and get some testing time in when I get back.

Signature · 1 line
Jens Axboe