dolfin team mailing list archive

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

Matthew Knepley wrote:
> On Mon, Jul 21, 2008 at 3:50 PM, Garth N. Wells <gnw20@xxxxxxxxx> wrote:
> > Anders Logg wrote:
> > > On Mon, Jul 21, 2008 at 01:48:23PM +0100, Garth N. Wells wrote:
> > > > Anders Logg wrote:
> > > > > I have updated the assembly benchmark to include also MTL4, see
> > > > >
> > > > >    bench/fem/assembly/
> > > > >
> > > > > Here are the current results:
> > > > >
> > > > > Assembly benchmark  |  Elasticity3D  PoissonP1  PoissonP2  PoissonP3  THStokes2D  NSEMomentum3D  StabStokes2D
> > > > > -------------------------------------------------------------------------------------------------------------
> > > > > uBLAS               |        9.0789    0.45645     3.8042     8.0736  14.937         9.2507        3.8455
> > > > > PETSc               |        7.7758    0.42798     3.5483     7.3898  13.945         8.1632         3.258
> > > > > Epetra              |        8.9516    0.45448     3.7976     8.0679  15.404         9.2341        3.8332
> > > > > MTL4                |        8.9729    0.45554     3.7966     8.0759  14.94          9.2568        3.8658
> > > > > Assembly            |         7.474    0.43673     3.7341     8.3793  14.633         7.6695        3.3878
> >
> > I specified in MTL4Matrix maximum 30 nonzeroes per row, and the results
> > change quite a bit,
> >
> >  Assembly benchmark  |  Elasticity3D  PoissonP1  PoissonP2  PoissonP3
> > THStokes2D  NSEMomentum3D  StabStokes2D
> >
> > -------------------------------------------------------------------------------------------------------------
> >  uBLAS               |        7.1881    0.32748     2.7633     5.8311
> >     10.968         7.0735        2.8184
> >  PETSc               |        5.7868    0.30673     2.5489     5.2344
> >     9.8896          6.069        2.3661
> >  MTL4                |        2.8641    0.18339     1.6628     2.6811
> >     2.8519         3.4843       0.85029
> >  Assembly            |        5.5564    0.30896     2.6858     5.9675
> >     10.622         5.7144        2.4519
> >
> >
> > MTL4 is a lot faster in all cases.
>
> Now I don't believe the numbers. If you preallocate, we do not do any
> extra processing
> outside of sorting the column indices (which every format must do for
> efficient operations).
> Thus, how would you save any time? If these are all in seconds, I will
> run a 2D Poisson here
> and tell you what I get. It would help to specify sizes with this benchmark :)

Take a look at bench/fem/assembly/ for the details.

Garth

>   Matt
>
> > Garth
> >
> >
> >
> > > > How was the MTL4 matrix intialised? I don't know if it does anything
> > > > with the sparsity pattern yet. I've been intialising MTL4 matrices by
> > > > hand so far with a guess as to the max number of nonzeroes per row.
> > > > Without setting this, the performance is near idenetical to uBLAS. When
> > > > it is set, I observe at least a factor two speed up.
> > > >
> > > > Garth
> > > The same way as all other backends, which is by a precomputed
> > > sparsity pattern. It looks like this is currently ignored in the
> > > MTL4Matrix implementation:
> > >
> > > void MTL4Matrix::init(const GenericSparsityPattern& sparsity_pattern)
> > > {
> > >   init(sparsity_pattern.size(0), sparsity_pattern.size(1));
> > > }
> > >
> > >
> > >
> > > ------------------------------------------------------------------------
> > >
> > > _______________________________________________
> > > DOLFIN-dev mailing list
> > > DOLFIN-dev@xxxxxxxxxx
> > > http://www.fenics.org/mailman/listinfo/dolfin-dev
> >
> > _______________________________________________
> > DOLFIN-dev mailing list
> > DOLFIN-dev@xxxxxxxxxx
> > http://www.fenics.org/mailman/listinfo/dolfin-dev