[Dune] dune-grid-howto examples programs on parallel computers

Carsten Gräser graeser at math.fu-berlin.de
Mon Sep 13 17:15:20 CEST 2010


Am 13.09.2010 17:02, schrieb Ankur Jayesh:
> Dear Sir,
> 
> I feel that using the dune sub-grid module would be beneficial for me.
> However, I understand that it is supported only for version 1.2.
> Should this be a reason for me to use version 1.2 rather than version 2.0?
> Is the module incorporated within the grid module now?
> The DUNE-fem module requires that DUNE be at version 1.2.1. In this case,
> would subgrid module work there?

Hello,
dune-subgrid is not part of the official dune modules.
Although there is no recent tarball on the dune-subgrid
homepage it works well with the 2.0 release and the trunk.
We will soon put an up-to-date tarball on the homepage
(http://numerik.mi.fu-berlin.de/dune-subgrid).

Until then I will send you a tarball in a separate e-mail.

Best regards
Carsten

> Regards
> Jayesh Vinay Badwaik
> 
> 
> 
> On Sat, Sep 4, 2010 at 12:43 PM, Ankur Jayesh <jayesh.ankur at gmail.com>wrote:
> 
>> Dear Sir,
>>
>>
>>
>>
>> Regards
>>
>>
>> On Wed, Sep 1, 2010 at 10:45 PM, Ankur Jayesh <jayesh.ankur at gmail.com>wrote:
>>
>>> Thank You for your information.
>>>
>>> Regards
>>> Jayesh Vinay Badwaik
>>>
>>>
>>> On 9/1/10, Markus Blatt <Markus.Blatt at iwr.uni-heidelberg.de> wrote:
>>>> Hi,
>>>>
>>>> On Wed, Sep 01, 2010 at 09:12:36PM +0530, Ankur Jayesh wrote:
>>>>> Dear Sir,
>>>>> I'm trying to use DUNE library v2.0 for my work. I am working on a
>>> cluster
>>>>> computer with MPICH2.
>>>>> The tests are not showing any speed-up due to the cluster. What could
>>> be
>>>>> the
>>>>> problem?
>>>>>
>>>>> The programs I've run are:
>>>>> gettingstarted
>>>>> finitevolume
>>>>
>>>> Sorry for the misleading MPI initializers in these sample programs.
>>>> Both of these programs are not parallel. If you run them on a cluster
>>>> every process computes the same sequential program.
>>>> For parallel programs you have to proceed to chapter 9.
>>>>
>>>> Cheers,
>>>>
>>>> Markus
>>>>
>>>> --
>>>> Markus Blatt, Interdisciplinary Center for Scientific Computing,
>>>> University Heidelberg, Im Neuenheimer Feld 368, D-69120 Heidelberg
>>>>
>>>> 'Mathematics knows no races or geographic boundaries; for
>>>> mathematics, the cultural world is one country' - David Hilbert




More information about the Dune mailing list