[mcstas-users] using 'icc' in 'oneAPI'

[Peter Kjær Willendrup]

Hi Uli,


I also heard good things about Intel oneAPI, but was not really animated to try it out until now.
(On the other hand I also don’t have any boxes with FPGA’s or Intel GPU’s, which in my understanding is where oneAPI starts to make the most sense.)


Using the default CFLAGS it certainly does not seem worth the effort, at least on macOS, here is a table with results from MPI-runs on 3 configurations:

1) clang + MPI on my Intel i7 2,7Ghz
2) oneAPI / icc + MPI on my Intel i7 2,7Ghz
3) clang + MPI on my fresh Apple M1

-> http://tmp.mcstas.org/3.0-dev-test/

As I write this I am running the same test on a Linux box with gcc and oneAPI, these will be uploaded to the same URL later - after which I will look into further optim flags etc.

- In the past I did recommend to use icc in place of gcc as it used to give a factor of 2 in performance, but it seems Open Source stroke back in the form of llvm/clang...


Best,
Peter


On 26 Apr 2021, at 16.10, Wildgruber, Christoph U. <wildgrubercu at ornl.gov<mailto:wildgrubercu at ornl.gov>> wrote:

Hi everybody,

I justy started learning a little bit about intel’s oneAPI.

My fist little project was top recompile one of my more demanding instruments with icc
on my (intel) laptop. That went well, but I was disappointed about the code performance
using the default options for icc which is no big surprise.

I am looking for suggestions what compiler switches I should consider to use.
Of course I am also wondering what other mcstas users experienced with respect to icc
in oneAPI..


Thanks a bunch,

Uli


P.S. No GPU’s involved at this point, I am just starting :-)
_______________________________________________
mcstas-users mailing list
mcstas-users at mcstas.org<mailto:mcstas-users at mcstas.org>
https://mailman2.mcstas.org/mailman/listinfo/mcstas-users



Peter Kjær Willendrup
Forskningsingeniør, Specialkonsulent
Næstformand for DTU Fysik LSU



DTU Physics

[cid:image001.gif at 01CCCAF1.5E6331F0]

Technical University of Denmark


[cid:image002.gif at 01CCCAF1.5E6331F0]






Department of Physics
Fysikvej
Building 307
DK-2800 Kongens Lyngby
Direct +45 2125 4612
Mobil +45 2125 4612
Fax +45 4593 2399
pkwi at fysik.dtu.dk<mailto:pkwi at fysik.dtu.dk>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20210428/d6169b10/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 58 bytes
Desc: image001.gif
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20210428/d6169b10/attachment-0002.gif>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image002.gif
Type: image/gif
Size: 1055 bytes
Desc: image002.gif
URL: <http://mailman2.mcstas.org/pipermail/mcstas-users/attachments/20210428/d6169b10/attachment-0003.gif>