Re: [patch 0/8]: use runnable load avg in balance

2013-05-29 Thread Alex Shi
On 05/28/2013 09:31 PM, Alex Shi wrote: >> > Tested the latest patch set (new 3/8 and 6/8) with pgbench, tip >> > 3.10.0-rc1 and 12 cpu X86 box, works well and still benefit ;-) > > Paul: > > Would you like to give more comments/ideas of this patch set? Peter, If no more idea of the blocked_loa

Re: [patch 0/8]: use runnable load avg in balance

2013-05-28 Thread Alex Shi
On 05/16/2013 03:29 PM, Michael Wang wrote: >> > This version changed the runnable load avg value setting for new task >> > in patch 3rd. >> > >> > We also tried to include blocked load avg in balance. but find many >> > benchmark >> > performance dropping. Guess the too bigger cpu load drive tas

Re: [patch 0/8]: use runnable load avg in balance

2013-05-16 Thread Alex Shi
On 05/16/2013 03:29 PM, Michael Wang wrote: >> > Michael Wang had tested previous version on pgbench on his box: >> > https://lkml.org/lkml/2013/4/2/1022 > Tested the latest patch set (new 3/8 and 6/8) with pgbench, tip > 3.10.0-rc1 and 12 cpu X86 box, works well and still benefit ;-) Thanks Mich

Re: [patch 0/8]: use runnable load avg in balance

2013-05-16 Thread Michael Wang
On 05/10/2013 11:17 PM, Alex Shi wrote: > This patchset bases on tip/sched/core. > > This version changed the runnable load avg value setting for new task > in patch 3rd. > > We also tried to include blocked load avg in balance. but find many benchmark > performance dropping. Guess the too bigger

Re: [patch 0/8]: use runnable load avg in balance

2013-05-14 Thread Alex Shi
On 05/14/2013 05:34 PM, Paul Turner wrote: >> > >> > We also tried to include blocked load avg in balance. but find many >> > benchmark >> > performance dropping. Guess the too bigger cpu load drive task to be waken >> > on remote CPU, and cause wrong decision in periodic balance. > Fundamentally,

Re: [patch 0/8]: use runnable load avg in balance

2013-05-14 Thread Paul Turner
On Fri, May 10, 2013 at 8:17 AM, Alex Shi wrote: > This patchset bases on tip/sched/core. > > This version changed the runnable load avg value setting for new task > in patch 3rd. > > We also tried to include blocked load avg in balance. but find many benchmark > performance dropping. Guess the to

Re: [patch 0/8]: use runnable load avg in balance

2013-05-14 Thread Alex Shi
On 05/10/2013 11:17 PM, Alex Shi wrote: > On SNB EP 4 sockets machine, the hackbench increased about 50%, and result > become stable. on other machines, hackbench increased about 2~10%. > oltp increased about 30% in NHM EX box. > netperf loopback also increased on SNB EP 4 sockets box. > no clear c

[patch 0/8]: use runnable load avg in balance

2013-05-10 Thread Alex Shi
This patchset bases on tip/sched/core. This version changed the runnable load avg value setting for new task in patch 3rd. We also tried to include blocked load avg in balance. but find many benchmark performance dropping. Guess the too bigger cpu load drive task to be waken on remote CPU, and ca