public inbox for gcc@gcc.gnu.org
 help / color / mirror / Atom feed
* [GSoC'19, libgomp work stealing] baseline benchmark results
       [not found] <3bad2d72c94412edadc8edf849279a7@cweb005.nm.nfra.io>
@ 2019-05-30  0:53 ` 김규래
  2019-05-30  9:11   ` Martin Jambor
  0 siblings, 1 reply; 3+ messages in thread
From: 김규래 @ 2019-05-30  0:53 UTC (permalink / raw)
  To: gcc; +Cc: jakub

​Hi everyone,
Just wanted to share some quick baseline benchmark results [3].
I ran LU decomposition on a AMD Ryzen Threadripper 1950x 16C/32T system.
LAPACK is currently plain loop parallel BLAS as far I believe.
And the upstream version of PLASMA uses OpenMP tasks [1].
The colored region is the 95% confidence interval.
Task parallelism seems to scale pretty well on such a small scale benchmark.
I hope work-stealing could improve this even more.
An
 
Ray Kim
 
[1] YarKhan, Asim, et al. "Porting the PLASMA numerical library to the OpenMP standard." International Journal of Parallel Programming 45.3 (2017): 612-633.
[2] https://bitbucket.org/icl/plasma/src/default/
[3] Link to benchmark plot: https://m.imgur.com/ysxs5ol
 

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [GSoC'19, libgomp work stealing] baseline benchmark results
  2019-05-30  0:53 ` [GSoC'19, libgomp work stealing] baseline benchmark results 김규래
@ 2019-05-30  9:11   ` Martin Jambor
  2019-05-30  9:15     ` 김규래
  0 siblings, 1 reply; 3+ messages in thread
From: Martin Jambor @ 2019-05-30  9:11 UTC (permalink / raw)
  To: 김규래, gcc; +Cc: jakub

Hi,

On Thu, May 30 2019, 김규래 wrote:
> ​Hi everyone,
> Just wanted to share some quick baseline benchmark results [3].
> I ran LU decomposition on a AMD Ryzen Threadripper 1950x 16C/32T system.
> LAPACK is currently plain loop parallel BLAS as far I believe.
> And the upstream version of PLASMA uses OpenMP tasks [1].
> The colored region is the 95% confidence interval.
> Task parallelism seems to scale pretty well on such a small scale benchmark.
> I hope work-stealing could improve this even more.
> An
>  
> Ray Kim
>  
> [1] YarKhan, Asim, et al. "Porting the PLASMA numerical library to the OpenMP standard." International Journal of Parallel Programming 45.3 (2017): 612-633.
> [2] https://bitbucket.org/icl/plasma/src/default/
> [3] Link to benchmark plot: https://m.imgur.com/ysxs5ol

This last link leads to some sort of Error 404 page of imgur, or at
least that's how I understand it.  Maybe you have not copy'n'pasted the
whole address?

Thanks,

Martin

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [GSoC'19, libgomp work stealing] baseline benchmark results
  2019-05-30  9:11   ` Martin Jambor
@ 2019-05-30  9:15     ` 김규래
  0 siblings, 0 replies; 3+ messages in thread
From: 김규래 @ 2019-05-30  9:15 UTC (permalink / raw)
  To: Martin Jambor, gcc; +Cc: jakub

Hi,
My bad.
I broke the link capitalization by mistake.
This is the correct link:
http://imgur.com/YsxS5Ol
​
Ray Kim
​
-----Original Message-----
From: "Martin Jambor"<mjambor@suse.cz>
To: "김규래"<msca8h@naver.com>; <gcc@gcc.gnu.org>;
Cc: <jakub@redhat.com>;
Sent: 2019-05-30 (목) 18:11:31 (GMT+09:00)
Subject: Re: [GSoC'19, libgomp work stealing] baseline benchmark results

Hi,

On Thu, May 30 2019, 김규래 wrote:
> ​Hi everyone,
> Just wanted to share some quick baseline benchmark results [3].
> I ran LU decomposition on a AMD Ryzen Threadripper 1950x 16C/32T system.
> LAPACK is currently plain loop parallel BLAS as far I believe.
> And the upstream version of PLASMA uses OpenMP tasks [1].
> The colored region is the 95% confidence interval.
> Task parallelism seems to scale pretty well on such a small scale benchmark.
> I hope work-stealing could improve this even more.
> An
>  
> Ray Kim
>  
> [1] YarKhan, Asim, et al. "Porting the PLASMA numerical library to the OpenMP standard." International Journal of Parallel Programming 45.3 (2017): 612-633.
> [2] https://bitbucket.org/icl/plasma/src/default/
> [3] Link to benchmark plot: https://m.imgur.com/ysxs5ol

This last link leads to some sort of Error 404 page of imgur, or at
least that's how I understand it.  Maybe you have not copy'n'pasted the
whole address?

Thanks,

Martin

​
​

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2019-05-30  9:15 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
     [not found] <3bad2d72c94412edadc8edf849279a7@cweb005.nm.nfra.io>
2019-05-30  0:53 ` [GSoC'19, libgomp work stealing] baseline benchmark results 김규래
2019-05-30  9:11   ` Martin Jambor
2019-05-30  9:15     ` 김규래

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).