[Trilinos-Users] benchmarking against ML and MueLU (help required about "optimal" settings)

[Riccardo Rossi]

Dear List,


we are in the process of publishing some benchmarks of a new linear solver
library (a prerelease version of the article is here
https://arxiv.org/pdf/1710.03940.pdf).

In order to get a reference performance baseline, we considered two
testcases and we obtained timings using Trilinos ML preconditioners. In all
cases we employed the solver defaults which we considered more relevant.


The two test cases considered are:

- 7 point laplacian testcase – here we use the “SA” default settings

- Unstructured Monolithic Navier-Stokes problem, in which velocity and
pressure are grouped in aggregates of size 4. We solve this by employing
ML’s “NSSA” settings.


Our finding is that ML is very fast but not very scalable.


Benchmark code and results can be found here


http://amgcl.readthedocs.io/en/latest/benchmarks.html


The article reviewers are considering that this comparison is not fair to
Trilinos (sorry, that was not our intention, we assumed that particularly
for the laplacian the SA settings were already finely tuned), and suggested
us to ask to Trilinos User community for better settings in order to
achieve a higher scalability.


Another suggestion is to try out the MueLU library instead, since the code
basis is expected to be more modern. To do so we would also need some
suggestions on the settings that should be employed for such a comparison.


Note that in either case we would like some settings which are still
reasonable defaults.


Any help on this would be highly appreciated


best regards

Denis & Riccardo


-- 


*Riccardo Rossi*

PhD, Civil Engineer


member of the Kratos Team: www.cimne.com/kratos

Tenure Track Lecturer at Universitat Politècnica de Catalunya,
BarcelonaTech (UPC)

Full Research Professor at International Center for Numerical Methods in
Engineering (CIMNE)


C/ Gran Capità, s/n, Campus Nord UPC, Ed. B0, Despatx 102

(please deliver post to the porters of building C1)

08034 – Barcelona – Spain – www.cimne.com  -

T.(+34) 93 401 56 96 skype: *rougered4*



<http://www.cimne.com/>

<https://www.facebook.com/cimne> <http://blog.cimne.com/>
<http://vimeo.com/cimne> <http://www.youtube.com/user/CIMNEvideos>
<http://www.linkedin.com/company/cimne> <https://twitter.com/cimne>

Les dades personals contingudes en aquest missatge són tractades amb la
finalitat de mantenir el contacte professional entre CIMNE i voste. Podra
exercir els drets d'accés, rectificació, cancel·lació i oposició,
dirigint-se a cimne at cimne.upc.edu. La utilització de la seva adreça de
correu electronic per part de CIMNE queda subjecte a les disposicions de la
Llei 34/2002, de Serveis de la Societat de la Informació i el Comerç
Electronic.

 Imprimiu aquest missatge, només si és estrictament necessari.
<http://www.cimne.com/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://trilinos.org/pipermail/trilinos-users/attachments/20180320/82079f03/attachment.html>