[announce] linux kernel performance project launch at sourceforge.net

[announce] linux kernel performance project launch at sourceforge.net

[Chen, Kenneth W]

I'm pleased to announce that we have established a linux kernel
performance project, hosted at sourceforge.net:

http://kernel-perf.sourceforge.net

As much as discussed various time in the past on LKML that Linux
kernel needs a systematic and disciplined way to measure and track
kernel performance on a regular basis.  To do that, we decided to
run a large set of benchmarks covering core components of the Linux
kernel (virtual memory management, I/O subsystem, process scheduler,
file system, network, device driver, etc) on a regular basis.
Benchmarks are run on a variety of platforms (4P Intel Xeon processor,
2P Xeon, several ia64 server boxes etc) every week, measuring the
latest snapshot of Linus' git development tree. Comprehensive performance
data from our tests will be published for easy access.

Our goal is to work with the Linux community to further enhance the
performance of the Linux kernel. The data available on the site allows
community members to closely track performance gains and losses with
every version of the kernel. Ultimately, we hope that this data will
result in performance increases in Linux kernel development.

The benchmark result pages are populated with a few benchmarks at the
moment.  In the coming weeks, we will be populating more benchmark data.
Happy surfing and hacking!!


		Ken Chen <kenneth.w.chen@intel.com>
		Intel Open Source Technology Center

Re: [announce] linux kernel performance project launch at sourceforge.net

[Andi Kleen]

"Chen, Kenneth W" <kenneth.w.chen@intel.com> writes:

> I'm pleased to announce that we have established a linux kernel
> performance project, hosted at sourceforge.net:
> 
> http://kernel-perf.sourceforge.net

That's very cool. Thanks a lot.

Would it be possible to add 2.4.30 numbers and perhaps one or two 
distro kernels (let's say RHEL3/4, SLES8/9) to the graphs 
as data points for comparison? These are all very tuned
kernels and would show where mainline is worse than them.

Also how did you run netperf? Locally or to some other machine? 
Perhaps that should be documented.

Some oprofile listings from a few of the test runs would be also nice.

-Andi

RE: [announce] linux kernel performance project launch at sourceforge.net

[Chen, Kenneth W]

ak@suse.de wrote on Thursday, July 14, 2005 3:18 PM
> "Chen, Kenneth W" <kenneth.w.chen@intel.com> writes:
> > I'm pleased to announce that we have established a linux kernel
> > performance project, hosted at sourceforge.net:
> > 
> > http://kernel-perf.sourceforge.net
> 
> That's very cool. Thanks a lot.

Thank you.


> Would it be possible to add 2.4.30 numbers and perhaps one or two 
> distro kernels (let's say RHEL3/4, SLES8/9) to the graphs 
> as data points for comparison? These are all very tuned
> kernels and would show where mainline is worse than them.

We did have a distro kernel in the graph originally and later decided
to go with pure mainline kernels for consistency.  We will see what we
can do to add them in the future.


> Also how did you run netperf? Locally or to some other machine? 
> Perhaps that should be documented.

Yes, that was netperf, running locally.  Thanks for the suggestion, I
will document that appropriately.


> Some oprofile listings from a few of the test runs would be also nice.

That is in the works.  We will upload profile data.  I'm having problem
with oprofile on some versions of kernel and that is being investigated
right now.

- Ken

Re: [announce] linux kernel performance project launch at sourceforge.net

[Andi Kleen]

> > Some oprofile listings from a few of the test runs would be also nice.
> 
> That is in the works.  We will upload profile data.  I'm having problem
> with oprofile on some versions of kernel and that is being investigated
> right now.

If you run statically compiled kernels you could as well use the
old style readprofile.  It just doesn't work with modules.

-Andi

RE: [announce] linux kernel performance project launch at sourceforge.net

[Chen, Kenneth W]

Alexey Dobriyan wrote on Thursday, July 14, 2005 3:34 PM
> On Friday 15 July 2005 00:21, Chen, Kenneth W wrote:
> > I'm pleased to announce that we have established a linux kernel
> > performance project, hosted at sourceforge.net:
> > 
> > http://kernel-perf.sourceforge.net
> 
> Perhaps, some cool-looking graphs instead of tables. Or you can write
> in red numbers where left deriative is smaller than zero. ;-)

I think sourceforge is being pounded hard by all the friendly kernel
hackers hitting the kernel performance page :-)  The site is just being
slow.  There are both tables and color charts in the result pages.

- Ken

Re: [announce] linux kernel performance project launch at sourceforge.net

[Alexey Dobriyan]

On Friday 15 July 2005 00:21, Chen, Kenneth W wrote:
> I'm pleased to announce that we have established a linux kernel
> performance project, hosted at sourceforge.net:
> 
> http://kernel-perf.sourceforge.net

Perhaps, some cool-looking graphs instead of tables. Or you can write in red
numbers where left deriative is smaller than zero. ;-)

> Comprehensive performance data from our tests will be published for easy
> access. 

Great! No, really. This means statistical errors.

Re: [announce] linux kernel performance project launch at sourceforge.net

[randy_dunlap]

On Fri, 15 Jul 2005 00:27:48 +0200 Andi Kleen wrote:

> > > Some oprofile listings from a few of the test runs would be also nice.
> > 
> > That is in the works.  We will upload profile data.  I'm having problem
> > with oprofile on some versions of kernel and that is being investigated
> > right now.
> 
> If you run statically compiled kernels you could as well use the
> old style readprofile.  It just doesn't work with modules.

It can be made to work with modules (and has been)[1],
but I'd just stick with not using modules, given a choice.

---
~Randy

[1] http://developer.osdl.org/rddunlap/modprofile/
(against 2.6.6)

Re: [announce] linux kernel performance project launch at sourceforge.net

[Paolo Ciarrocchi]

2005/7/14, Chen, Kenneth W <kenneth.w.chen@intel.com>:
> I'm pleased to announce that we have established a linux kernel
> performance project, hosted at sourceforge.net:
> 
> http://kernel-perf.sourceforge.net

The web site is still very slow (@10:00 CET)  but the your work is really cool!

Do you think it's possbile to somehow merge your work toghether with:
http://ftp.kernel.org/pub/linux/kernel/people/mbligh/abat

And, do you think it's possible and usefull to add this test:
http://marc.theaimsgroup.com/?l=linux-kernel&m=112140025310254&w=2

Thanks!!
-- 
http://paoloc.blogspot.com

RE: [announce] linux kernel performance project launch at sourceforge.net

[Al Boldi]

Ken Chen <kenneth.w.chen@intel.com> wrote: {
...Linux kernel needs a systematic and disciplined way to measure and track
kernel performance on a regular basis.
}

Nice, but the numbers would have more meaning if they were put in relation
to System Load ( CPU,MEM,DISK,NET,... )

Also, a test that saturates resources would easily expose deeper Kernel
problems.

Thanks,
		Al

[announce] linux kernel performance project launch at sourceforge.net

[Chen, Kenneth W]

Forward for people who might missed the "linux kernel performance"
announcement on LKML yesterday. It has performance data for ia64,
in case if you are interested, you are welcome to browse the data
and join the discussion on kernel performance.

		Ken Chen <kenneth.w.chen@intel.com>
		Intel Open Source Technology Center



-----Original Message-----
From: Chen, Kenneth W [mailto:kenneth.w.chen@intel.com] 
Sent: Thursday, July 14, 2005 1:21 PM
To: linux-kernel@vger.kernel.org
Subject: [announce] linux kernel performance project launch at sourceforge.net

I'm pleased to announce that we have established a linux kernel
performance project, hosted at sourceforge.net:

http://kernel-perf.sourceforge.net

As much as discussed various time in the past on LKML that Linux
kernel needs a systematic and disciplined way to measure and track
kernel performance on a regular basis.  To do that, we decided to
run a large set of benchmarks covering core components of the Linux
kernel (virtual memory management, I/O subsystem, process scheduler,
file system, network, device driver, etc) on a regular basis.
Benchmarks are run on a variety of platforms (4P Intel Xeon processor,
2P Xeon, several ia64 server boxes etc) every week, measuring the
latest snapshot of Linus' git development tree. Comprehensive performance
data from our tests will be published for easy access.

Our goal is to work with the Linux community to further enhance the
performance of the Linux kernel. The data available on the site allows
community members to closely track performance gains and losses with
every version of the kernel. Ultimately, we hope that this data will
result in performance increases in Linux kernel development.

The benchmark result pages are populated with a few benchmarks at the
moment.  In the coming weeks, we will be populating more benchmark data.
Happy surfing and hacking!!


		Ken Chen <kenneth.w.chen@intel.com>
		Intel Open Source Technology Center