Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-27 Thread Frederic Weisbecker
On Thu, Aug 27, 2015 at 05:09:21PM +0200, Thomas Gleixner wrote: > On Thu, 27 Aug 2015, Steven Rostedt wrote: > > On Thu, 27 Aug 2015 15:18:49 +0200 > > Frederic Weisbecker wrote: > > > > > On Wed, Aug 26, 2015 at 04:45:44PM -0700, Hideaki Kimura wrote: > > > > I totally agree that this is not a

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-27 Thread Thomas Gleixner
On Thu, 27 Aug 2015, Steven Rostedt wrote: > On Thu, 27 Aug 2015 15:18:49 +0200 > Frederic Weisbecker wrote: > > > On Wed, Aug 26, 2015 at 04:45:44PM -0700, Hideaki Kimura wrote: > > > I totally agree that this is not a perfect solution. If there are 10x more > > > cores and sockets, just the ato

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-27 Thread Steven Rostedt
On Thu, 27 Aug 2015 15:18:49 +0200 Frederic Weisbecker wrote: > On Wed, Aug 26, 2015 at 04:45:44PM -0700, Hideaki Kimura wrote: > > I totally agree that this is not a perfect solution. If there are 10x more > > cores and sockets, just the atomic fetch_add might be too expensive. > > > > However,

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-27 Thread Frederic Weisbecker
On Wed, Aug 26, 2015 at 04:45:44PM -0700, Hideaki Kimura wrote: > I totally agree that this is not a perfect solution. If there are 10x more > cores and sockets, just the atomic fetch_add might be too expensive. > > However, it's comparatively/realistically the best thing we can do without > any d

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Hideaki Kimura
On 08/26/2015 04:13 PM, Frederic Weisbecker wrote: On Wed, Aug 26, 2015 at 03:53:26PM -0700, Hideaki Kimura wrote: Sure, let me elaborate. Executive summary: Yes, enabling a process-wide timer in such a large machine is not wise, but sometimes users/applications cannot avoid it. The issue

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Frederic Weisbecker
On Wed, Aug 26, 2015 at 03:53:26PM -0700, Hideaki Kimura wrote: > Sure, let me elaborate. > > Executive summary: > Yes, enabling a process-wide timer in such a large machine is not wise, but > sometimes users/applications cannot avoid it. > > > The issue was observed actually not in a database

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Hideaki Kimura
Sure, let me elaborate. Executive summary: Yes, enabling a process-wide timer in such a large machine is not wise, but sometimes users/applications cannot avoid it. The issue was observed actually not in a database itself but in a common library it links to; gperftools. The database itsel

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Jason Low
On Wed, 2015-08-26 at 19:08 +0200, Oleg Nesterov wrote: > On 08/26, Jason Low wrote: > > > > Hi Andrew, > > > > On Tue, 2015-08-25 at 20:27 -0700, Andrew Morton wrote: > > > On Tue, 25 Aug 2015 20:17:45 -0700 Jason Low wrote: > > > > > > > When running a database workload on a 16 socket machine, t

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Oleg Nesterov
On 08/26, Jason Low wrote: > > Hi Andrew, > > On Tue, 2015-08-25 at 20:27 -0700, Andrew Morton wrote: > > On Tue, 25 Aug 2015 20:17:45 -0700 Jason Low wrote: > > > > > When running a database workload on a 16 socket machine, there were > > > scalability issues related to itimers. > > > > > > Commi

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-26 Thread Jason Low
Hi Andrew, On Tue, 2015-08-25 at 20:27 -0700, Andrew Morton wrote: > On Tue, 25 Aug 2015 20:17:45 -0700 Jason Low wrote: > > > When running a database workload on a 16 socket machine, there were > > scalability issues related to itimers. > > > > Commit 1018016c706f addressed the issue with the

Re: [PATCH 0/3] timer: Improve itimers scalability

2015-08-25 Thread Andrew Morton
On Tue, 25 Aug 2015 20:17:45 -0700 Jason Low wrote: > When running a database workload on a 16 socket machine, there were > scalability issues related to itimers. > > Commit 1018016c706f addressed the issue with the thread_group_cputimer > spinlock taking up a significant portion of total run ti

[PATCH 0/3] timer: Improve itimers scalability

2015-08-25 Thread Jason Low
When running a database workload on a 16 socket machine, there were scalability issues related to itimers. Commit 1018016c706f addressed the issue with the thread_group_cputimer spinlock taking up a significant portion of total run time. This patch series address the other issue where a lot of ti