dolfin team mailing list archive

["Garth N. Wells" <gnw20@xxxxxxxxx>]

Anders Logg wrote:
> On Tue, Sep 15, 2009 at 11:20:15AM +0200, Garth N. Wells wrote:
> > Anders Logg wrote:
> > > On Mon, Sep 14, 2009 at 11:57:19PM +0200, Anders Logg wrote:
> > > > On Thu, Sep 10, 2009 at 10:30:55PM +0200, Anders Logg wrote:
> > > >
> > > > > ok, so it just remains to
> > > > >
> > > > > 1. Check parallel speedup (I don't think anyone has dared to try this yet).
> > > > Does anyone have time/resources to make a quick/simple check of the
> > > > speedup we get for some simple test case? Johan? Niclas?
> > > >
> > > > > 2. Make all demos run or exit gracefully in parallel.
> > > > The current demos are still failing in parallel:
> > > >
> > > >   [**] demo/fem/simple/cpp
> > > >   [**] demo/fem/simple/python
> > > >   [**] demo/function/eval/cpp
> > > >   [**] demo/function/eval/python
> > > >   [**] demo/mesh/meshfunction/cpp --> Ola
> > > >   [**] demo/pde/mixed-poisson/cpp
> > > >   [**] demo/pde/mixed-poisson/python
> > > >
> > > > Please take a look and see if you know what to do.
> > > The interpolation bug and most of the above demos have been fixed so
> > > it remains just to
> > >
> > > 1. Check parallel speedup
> > >
> > > Does anyone have a current DOLFIN build on bigblue.simula.no? Hake?
> > I don't think that this is necessary for a release. We won't make any
> > changes to the code at this stage based on the results.
> >
> > Garth
>
> I think it would be good to have some confidence that we actually get
> some speedup. We won't make any changes but it will be good to know
> how good (or bad) it is. It should also be fairly easy to check.

Yes, but why should the release depend on this if we're not going to 
make changes?

Garth

> --
> Anders
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> DOLFIN-dev mailing list
> DOLFIN-dev@xxxxxxxxxx
> http://www.fenics.org/mailman/listinfo/dolfin-dev