Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-23 Thread Alex Shi
>> >> Maybe we can skip local group since it's a bottom-up search so we know >> there's no idle cpu in the lower domain from the prior iteration. >> > > I did this change but seems results are worse on my machines, guess start > seeking idle cpu bottom up is a bad idea. > The following is full

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-23 Thread Alex Shi
Maybe we can skip local group since it's a bottom-up search so we know there's no idle cpu in the lower domain from the prior iteration. I did this change but seems results are worse on my machines, guess start seeking idle cpu bottom up is a bad idea. The following is full version with

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
On 01/21/2013 10:40 AM, Preeti U Murthy wrote: > Hi Alex, > Thank you very much for running the below benchmark on > blocked_load+runnable_load:) Just a few queries. > > How did you do the wake up balancing? Did you iterate over the L3 > package looking for an idle cpu? Or did you just query the

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Preeti U Murthy
Hi Alex, Thank you very much for running the below benchmark on blocked_load+runnable_load:) Just a few queries. How did you do the wake up balancing? Did you iterate over the L3 package looking for an idle cpu? Or did you just query the L2 package for an idle cpu? I think when you are using

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
>>> The blocked load of a cluster will be high if the blocked tasks have >>> run recently. The contribution of a blocked task will be divided by 2 >>> each 32ms, so it means that a high blocked load will be made of recent >>> running tasks and the long sleeping tasks will not influence the load

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
On 01/09/2013 11:14 AM, Preeti U Murthy wrote: > Here comes the point of making both load balancing and wake up > balance(select_idle_sibling) co operative. How about we always schedule > the woken up task on the prev_cpu? This seems more sensible considering > load balancing

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
On 01/09/2013 11:14 AM, Preeti U Murthy wrote: Here comes the point of making both load balancing and wake up balance(select_idle_sibling) co operative. How about we always schedule the woken up task on the prev_cpu? This seems more sensible considering load balancing considers blocked load as

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
The blocked load of a cluster will be high if the blocked tasks have run recently. The contribution of a blocked task will be divided by 2 each 32ms, so it means that a high blocked load will be made of recent running tasks and the long sleeping tasks will not influence the load balancing.

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Preeti U Murthy
Hi Alex, Thank you very much for running the below benchmark on blocked_load+runnable_load:) Just a few queries. How did you do the wake up balancing? Did you iterate over the L3 package looking for an idle cpu? Or did you just query the L2 package for an idle cpu? I think when you are using

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-20 Thread Alex Shi
On 01/21/2013 10:40 AM, Preeti U Murthy wrote: Hi Alex, Thank you very much for running the below benchmark on blocked_load+runnable_load:) Just a few queries. How did you do the wake up balancing? Did you iterate over the L3 package looking for an idle cpu? Or did you just query the L2

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Alex Shi
On 01/17/2013 01:17 PM, Namhyung Kim wrote: > On Wed, 16 Jan 2013 22:08:21 +0800, Alex Shi wrote: >> On 01/08/2013 04:41 PM, Preeti U Murthy wrote: >>> Hi Mike, >>> >>> Thank you very much for such a clear and comprehensive explanation. >>> So when I put together the problem and the proposed

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Preeti U Murthy
Hi Namhyung, >> I re-written the patch as following. hackbench/aim9 doest show clean >> performance change. >> Actually we can get some profit. it also will be very slight. :) >> BTW, it still need another patch before apply this. Just to show the logical. >> >> === >>> From

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Preeti U Murthy
Hi Alex, On 01/16/2013 07:38 PM, Alex Shi wrote: > On 01/08/2013 04:41 PM, Preeti U Murthy wrote: >> Hi Mike, >> >> Thank you very much for such a clear and comprehensive explanation. >> So when I put together the problem and the proposed solution pieces in the >> current >> scheduler

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Preeti U Murthy
Hi Alex, On 01/16/2013 07:38 PM, Alex Shi wrote: On 01/08/2013 04:41 PM, Preeti U Murthy wrote: Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in the current scheduler scalability,the

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Preeti U Murthy
Hi Namhyung, I re-written the patch as following. hackbench/aim9 doest show clean performance change. Actually we can get some profit. it also will be very slight. :) BTW, it still need another patch before apply this. Just to show the logical. === From

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-17 Thread Alex Shi
On 01/17/2013 01:17 PM, Namhyung Kim wrote: On Wed, 16 Jan 2013 22:08:21 +0800, Alex Shi wrote: On 01/08/2013 04:41 PM, Preeti U Murthy wrote: Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-16 Thread Namhyung Kim
On Wed, 16 Jan 2013 22:08:21 +0800, Alex Shi wrote: > On 01/08/2013 04:41 PM, Preeti U Murthy wrote: >> Hi Mike, >> >> Thank you very much for such a clear and comprehensive explanation. >> So when I put together the problem and the proposed solution pieces in the >> current >> scheduler

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-16 Thread Alex Shi
On 01/08/2013 04:41 PM, Preeti U Murthy wrote: > Hi Mike, > > Thank you very much for such a clear and comprehensive explanation. > So when I put together the problem and the proposed solution pieces in the > current > scheduler scalability,the following was what I found: > > 1.

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-16 Thread Alex Shi
On 01/08/2013 04:41 PM, Preeti U Murthy wrote: Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in the current scheduler scalability,the following was what I found: 1. select_idle_sibling()

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-16 Thread Namhyung Kim
On Wed, 16 Jan 2013 22:08:21 +0800, Alex Shi wrote: On 01/08/2013 04:41 PM, Preeti U Murthy wrote: Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in the current scheduler scalability,the

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Preeti U Murthy
Here comes the point of making both load balancing and wake up balance(select_idle_sibling) co operative. How about we always schedule the woken up task on the prev_cpu? This seems more sensible considering load balancing considers blocked load as being a part of the load of

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Vincent Guittot
On 8 January 2013 07:06, Preeti U Murthy wrote: > On 01/07/2013 09:18 PM, Vincent Guittot wrote: >> On 2 January 2013 05:22, Preeti U Murthy wrote: >>> Hi everyone, >>> I have been looking at how different workloads react when the per entity >>> load tracking metric is integrated into the load

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Preeti U Murthy
Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in the current scheduler scalability,the following was what I found: 1. select_idle_sibling() is needed as an agent to correctly find the right cpu

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Preeti U Murthy
Hi Mike, Thank you very much for such a clear and comprehensive explanation. So when I put together the problem and the proposed solution pieces in the current scheduler scalability,the following was what I found: 1. select_idle_sibling() is needed as an agent to correctly find the right cpu

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Vincent Guittot
On 8 January 2013 07:06, Preeti U Murthy pre...@linux.vnet.ibm.com wrote: On 01/07/2013 09:18 PM, Vincent Guittot wrote: On 2 January 2013 05:22, Preeti U Murthy pre...@linux.vnet.ibm.com wrote: Hi everyone, I have been looking at how different workloads react when the per entity load

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-08 Thread Preeti U Murthy
Here comes the point of making both load balancing and wake up balance(select_idle_sibling) co operative. How about we always schedule the woken up task on the prev_cpu? This seems more sensible considering load balancing considers blocked load as being a part of the load of cpu2. Hi Preeti,

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-07 Thread Preeti U Murthy
On 01/07/2013 09:18 PM, Vincent Guittot wrote: > On 2 January 2013 05:22, Preeti U Murthy wrote: >> Hi everyone, >> I have been looking at how different workloads react when the per entity >> load tracking metric is integrated into the load balancer and what are >> the possible reasons for it. >>

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-07 Thread Vincent Guittot
On 2 January 2013 05:22, Preeti U Murthy wrote: > Hi everyone, > I have been looking at how different workloads react when the per entity > load tracking metric is integrated into the load balancer and what are > the possible reasons for it. > > I had posted the integration patch earlier: >

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-07 Thread Vincent Guittot
On 2 January 2013 05:22, Preeti U Murthy pre...@linux.vnet.ibm.com wrote: Hi everyone, I have been looking at how different workloads react when the per entity load tracking metric is integrated into the load balancer and what are the possible reasons for it. I had posted the integration

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-07 Thread Preeti U Murthy
On 01/07/2013 09:18 PM, Vincent Guittot wrote: On 2 January 2013 05:22, Preeti U Murthy pre...@linux.vnet.ibm.com wrote: Hi everyone, I have been looking at how different workloads react when the per entity load tracking metric is integrated into the load balancer and what are the possible

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Mike Galbraith
On Mon, 2013-01-07 at 10:59 +0530, Preeti U Murthy wrote: > Hi Mike, > Thank you very much for your inputs.Just a few thoughts so that we are > clear with the problems so far in the scheduler scalability and in what > direction we ought to move to correct them. > > 1. During fork or exec,the

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Preeti U Murthy
Hi Mike, Thank you very much for your inputs.Just a few thoughts so that we are clear with the problems so far in the scheduler scalability and in what direction we ought to move to correct them. 1. During fork or exec,the scheduler goes through find_idlest_group() and find_idlest_cpu() in

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Mike Galbraith
On Sat, 2013-01-05 at 09:13 +0100, Mike Galbraith wrote: > I still have a 2.6-rt problem I need to find time to squabble with, but > maybe I'll soonish see if what you did plus what I did combined works > out on that 4x10 core box where current is _so_ unbelievably horrible. > Heck, it can't get

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Mike Galbraith
On Sat, 2013-01-05 at 09:13 +0100, Mike Galbraith wrote: I still have a 2.6-rt problem I need to find time to squabble with, but maybe I'll soonish see if what you did plus what I did combined works out on that 4x10 core box where current is _so_ unbelievably horrible. Heck, it can't get any

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Preeti U Murthy
Hi Mike, Thank you very much for your inputs.Just a few thoughts so that we are clear with the problems so far in the scheduler scalability and in what direction we ought to move to correct them. 1. During fork or exec,the scheduler goes through find_idlest_group() and find_idlest_cpu() in

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-06 Thread Mike Galbraith
On Mon, 2013-01-07 at 10:59 +0530, Preeti U Murthy wrote: Hi Mike, Thank you very much for your inputs.Just a few thoughts so that we are clear with the problems so far in the scheduler scalability and in what direction we ought to move to correct them. 1. During fork or exec,the scheduler

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-05 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: > Subject: [PATCH] sched: Merge select_idle_sibling with the behaviour of > SD_BALANCE_WAKE > > The function of select_idle_sibling() is to place the woken up task in the > vicinity of the waking cpu or on the previous cpu depending on

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-05 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: Subject: [PATCH] sched: Merge select_idle_sibling with the behaviour of SD_BALANCE_WAKE The function of select_idle_sibling() is to place the woken up task in the vicinity of the waking cpu or on the previous cpu depending on what

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-04 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: > Hi Mike, > > Thank you very much for your feedback.Considering your suggestions,I have > posted out a > proposed solution to prevent select_idle_sibling() from becoming a > disadvantage to normal > load balancing,rather aiding it. >

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-04 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: Hi Mike, Thank you very much for your feedback.Considering your suggestions,I have posted out a proposed solution to prevent select_idle_sibling() from becoming a disadvantage to normal load balancing,rather aiding it. **This

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-03 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: > Hi Mike, > > Thank you very much for your feedback.Considering your suggestions,I have > posted out a > proposed solution to prevent select_idle_sibling() from becoming a > disadvantage to normal > load balancing,rather aiding it. >

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-03 Thread Preeti U Murthy
Hi Mike, Thank you very much for your feedback.Considering your suggestions,I have posted out a proposed solution to prevent select_idle_sibling() from becoming a disadvantage to normal load balancing,rather aiding it. **This patch is *without* the enablement of the per entity load tracking

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-03 Thread Preeti U Murthy
Hi Mike, Thank you very much for your feedback.Considering your suggestions,I have posted out a proposed solution to prevent select_idle_sibling() from becoming a disadvantage to normal load balancing,rather aiding it. **This patch is *without* the enablement of the per entity load tracking

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-03 Thread Mike Galbraith
On Thu, 2013-01-03 at 16:08 +0530, Preeti U Murthy wrote: Hi Mike, Thank you very much for your feedback.Considering your suggestions,I have posted out a proposed solution to prevent select_idle_sibling() from becoming a disadvantage to normal load balancing,rather aiding it. **This

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-02 Thread Mike Galbraith
On Wed, 2013-01-02 at 09:52 +0530, Preeti U Murthy wrote: > Hi everyone, > I have been looking at how different workloads react when the per entity > load tracking metric is integrated into the load balancer and what are > the possible reasons for it. > > I had posted the integration patch

Re: sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-02 Thread Mike Galbraith
On Wed, 2013-01-02 at 09:52 +0530, Preeti U Murthy wrote: Hi everyone, I have been looking at how different workloads react when the per entity load tracking metric is integrated into the load balancer and what are the possible reasons for it. I had posted the integration patch earlier:

sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-01 Thread Preeti U Murthy
Hi everyone, I have been looking at how different workloads react when the per entity load tracking metric is integrated into the load balancer and what are the possible reasons for it. I had posted the integration patch earlier: https://lkml.org/lkml/2012/11/15/391 Essentially what I am doing

sched: Consequences of integrating the Per Entity Load Tracking Metric into the Load Balancer

2013-01-01 Thread Preeti U Murthy
Hi everyone, I have been looking at how different workloads react when the per entity load tracking metric is integrated into the load balancer and what are the possible reasons for it. I had posted the integration patch earlier: https://lkml.org/lkml/2012/11/15/391 Essentially what I am doing