[petsc-dev] Zoltan 3.81

[Rhodri Davies]

Hi Matt,

Thanks for the update and sorry for my slow response. I’ve tested today and the new version of zoltan downloads and compiles fine through PETSC. However, we still require the fortran .mod files within Fluidity. Rather than add the new flag, as I previously suggested (which was zoltan specific: -with-zoltan-fortran-interfaces), could the fortran interfaces in zoltan (i.e. the configure flag “ -enable-f90interface”) simply be included when we specify -with-fortran-interfaces=1 in petsc configure?

Best wishes,

Rhod

---------------------------------------------------------------------------------------------------
ARC Future Fellow
Research School of Earth Sciences
The Australian National University
Canberra, ACT, 0200, Australia

T: +61 (0)2 6125 3643
E: rhodri.davies at anu.edu.au<mailto:rhodri.davies at anu.edu.au>
W: http://rses.anu.edu.au/people/dr-rhodri-davies
---------------------------------------------------------------------------------------------------



On 6 May 2015, at 06:04, Matthew Knepley <knepley at gmail.com<mailto:knepley at gmail.com>> wrote:

On Thu, Apr 30, 2015 at 10:43 AM, Michael Lange <michael.lange at imperial.ac.uk<mailto:michael.lange at imperial.ac.uk>> wrote:
On 30/04/15 14:19, Matthew Knepley wrote:
On Thu, Apr 30, 2015 at 3:47 PM, Rhodri Davies <rhodri.davies at anu.edu.au<mailto:rhodri.davies at anu.edu.au>> wrote:
Dear Devs,

Would it be possible for you to update the version of Zoltan packaged with PETSc to version 3.8.1? This version upgrades Zoltan's interface to ParMETIS v4 and Scotch v6, which are the versions of these libraries that PETSc currently defaults to.

I can do this when I get home. Note that we are not really using this in PETSc now, do you could just install it if you want.

On a related note, is there currently a way to pass the Zoltan configure flag “ -enable-f90interface", to the zoltan build, through PETSc? In Fluidity, we currently depend on these fortran interfaces (.mod files), so this option would be highly desirable for us. If it’s not currently possible, could you add an optional “--with-zoltan-fortran-interfaces” to PETSc configure?

Yes, this is easy to pass. Note that Michael Lange has recently done some great work on expanding
Fluidity to use DMPlex for PETSc for dynamic rebalancing. Have you talked to him? I think it will be
ready in Fluidity later this year.
Hi Matt,

Unfortunately, Zoltan is used in multiple places in Fluidity, so there will still be a dependency overlap for a while. A version update would thus be much appreciated.

I have bumped the version in 'next' and tried it out on my laptop. Let me know if it works for you guys.

  Thanks,

    Matt

Thanks,
Michael

  Thanks,

     Matt

Best wishes,

Rhodri

---------------------------------------------------------------------------------------------------
ARC Future Fellow
Research School of Earth Sciences
The Australian National University
Canberra, ACT, 0200, Australia

T: +61 (0)2 6125 3643<tel:%2B61%20%280%292%206125%203643>
E: rhodri.davies at anu.edu.au<mailto:rhodri.davies at anu.edu.au>
W: http://rses.anu.edu.au/people/dr-rhodri-davies
---------------------------------------------------------------------------------------------------






--
What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
-- Norbert Wiener




--
What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
-- Norbert Wiener

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20150510/2d2611ba/attachment.html>