sched: change nohz idle load balancing logic to push model
in [Kernel]
|
Prev: [tip:perf/core] perf tools: Add mode to build without newt support
Next: [PATCH 1/3] DMAENGINE: DMA40 support paused channel status
From: Nigel Cunningham on 17 May 2010 18:40 Hi Suresh. On 18/05/10 04:27, Suresh Siddha wrote: > This is an updated version of patchset which is posted earlier at > http://lkml.org/lkml/2009/12/10/470 > > Description: > Existing nohz idle load balance logic uses the pull model, with one > idle load balancer CPU nominated on any partially idle system and that > balancer CPU not going into nohz mode. With the periodic tick, the > balancer does the idle balancing on behalf of all the CPUs in nohz mode. > > This is not very optimal and has few issues: > * the balancer will continue to have periodic ticks and wakeup > frequently (HZ rate), even though it may not have any rebalancing to do on > behalf of any of the idle CPUs. > * On x86 and CPUs that have APIC timer stoppage on idle CPUs, this periodic > wakeup can result in an additional interrupt on a CPU doing the timer > broadcast. > > The alternative is to have a push model, where all idle CPUs can enter nohz > mode and any busy CPU kicks one of the idle CPUs to take care of idle > balancing on behalf of a group of idle CPUs. > > Following patches switches idle load balancer to this push approach. > > Updates from the previous version: > > * Busy CPU uses send_remote_softirq() for invoking SCHED_SOFTIRQ on the > idle load balancing cpu, which does the load balancing on behalf of > all the idle CPUs. > > * Dropped the per NUMA node nohz load balancing as it doesn't detect > certain imbalance scenarios. This will be addressed later. > > Signed-off-by: Suresh Siddha<suresh.b.siddha(a)intel.com> > Signed-off-by: Venkatesh Pallipadi<venki(a)google.com> Sounds great. I'm in the middle of packing at the moment, but will try to find time to test the patches. It might help to show the difference that the patch series makes (powertop output?) Regards, Nigel -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Nigel Cunningham on 20 May 2010 07:10
Hi. I finally found the time to give this patch series a try. Minor updates were required, but I'm now running it against 2.6.34. "Load balancing tick" is still number one in my powertop list of top causes of wakeups (sitting at ~60 to 80 per second as I type this, with ~170 wakeups per second total). Comparing this to the numbers I posted earlier, we seem to have a win. I do wonder, though, whether further work could still be done. If I take one core offline, for example, I'm still getting load balancing ticks. Intuitively, I'd expect there to be no need for them with only one core available. But maybe I'm just ignorant of what's going on. Regards, Nigel -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |