[gmx-developers] Use of SVE in GROMACS
guido.giuntoli at huawei.com
Wed Oct 7 16:28:09 CEST 2020
Thanks a lot for your prompt and accurate answer.
The non-bonded benchmark you suggested looks very convenient, as I can understand it correctly it does not require any data IO and the initial conditions of the problem are generated using the command line parameters, is that correct?
With this benchmark I see most of the computing time in nbnxm_kernel_ElecEw_VdwLJCombGeom_F_4xm which calls kernel_outer.h and this last to kernel_inner.h. The idea of implementing this through includes is for avoiding function call overhead?
I think this benchmark will simplify a lot the kernel extraction since the number of functions that are called are lesser and simpler. Thanks for this useful suggestion.
Best regards | Mit freundlichen Grüßen
Software Solution Architect for Advance Computing
HUAWEI TECHNOLOGIES Duesseldorf GmbH
Hansaallee 205, 40549 Dusseldorf, Germany, www.huawei.com
Registered Office: Düsseldorf, Register Court Düsseldorf, HRB 56063,
Managing Director: Li Peng, Li Jian, Shi Yanli
Sitz der Gesellschaft: Düsseldorf, Amtsgericht Düsseldorf, HRB 56063,
Geschäftsführer: Li Peng, Li Jian, Shi Yanli
This e-mail and its attachments contain confidential information from HUAWEI, which is intended only for the person or entity whose address is listed above. Any use of the information contained herein in any way (including, but not limited to, total or partial disclosure, reproduction, or dissemination) by persons other than the intended recipient(s) is prohibited. If you receive this e-mail in error, please notify the sender by phone or email immediately and delete it!
From: gromacs.org_gmx-developers-bounces at maillist.sys.kth.se [mailto:gromacs.org_gmx-developers-bounces at maillist.sys.kth.se] On Behalf Of Szilárd Páll
Sent: Wednesday, October 7, 2020 1:58 PM
To: Discussion list for GROMACS development <gmx-developers at gromacs.org>
Cc: Matthias Gries <matthias.gries at huawei.com>; gromacs.org_gmx-developers at maillist.sys.kth.se
Subject: Re: [gmx-developers] Use of SVE in GROMACS
Initial SVE support has just been merged a few days ago and will be included in the 2021 release:
This is initial support with a fixed implementation with some optimizations for A64fx.
Your contributions are welcome! For the 2021 release, which will have a beta release very soon, we can only accept minor improvements and tweaks. After the release-2021 branch is forked, further development should target the master branch.
See further replies below.
> 3) Is this routine responsible only of the non-bonded forces? Does the “nb” prefix indicate “non-bonded”?
> 4) Is this same effect (“do_nb_verlet” the most expensive) measured for others benchmarks?
Yes. There are two possible exceptions:
- strong scaling where communication and other less algorithms that scales less well than the nonbonded pair-interactions starts to dominate;
- uncommon rare simulation types where other very high cost algorithms are used.
The former should become clear when you scale out to a few hundred atoms/core. The latter I'd suggest to not worry about.
> 5) Are there others ready-to-use and small benchmarks like the “ion_channel.tpr” available (<150K atoms)? Test Case B from UEABS (3.3 M atoms) is not suitable for my hardware resources.
You can use the "rnase" benchamrk systems 19k/24k atoms (see here http://ftp.gromacs.org/benchmarks/).
You might also find useful the nonbonded-benchmark tool, see:
Let us know if you have further questions.
> Thanks a lot, Guido.
> Gromacs Developers mailing list
> * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
Gromacs Developers mailing list
* Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List before posting!
* Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
* For (un)subscribe requests visit
https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers or send a mail to gmx-developers-request at gromacs.org.
More information about the gromacs.org_gmx-developers