Hello all in PETSc team and others,<br><br>I tried to install/configure PETSc 2.3.3-p5 on my FC6 desktop computer. Following Happend:<br><br>[petsc1@localhost petsc-2.3.3-p5]$ ./config/configure.py --with-cc=gcc --with-fc=gfortran --with-cxx=g++ --download-f-blas-lapack=1 --download-mpich=1 --with-scalar-type=complex --with-clanguage=cxx
<br>=================================================================================<br> Configuring PETSc to compile on your system <br>=================================================================================
<br>TESTING: configureLibrary from config.packages.BlasLapack(python/BuildSystem/con*********************************************************************************<br> UNABLE to CONFIGURE with GIVEN OPTIONS (see
configure.log for details):<br>---------------------------------------------------------------------------------------<br>Error downloading fblaslapack.tar.gz requested with -with-f-blas-lapack option<br>*********************************************************************************
<br><br style="font-weight: bold; color: rgb(51, 51, 255);"><span style="font-weight: bold; color: rgb(51, 51, 255);">How to tackle this situation?<br><br><span style="font-weight: bold;">At the moment I want </span>to develop & run
<span style="text-decoration: underline;">PARALLEL</span> codes on my FC6 Desktop (single machine). <span style="font-weight: bold;"><span style="font-weight: bold;">Should I use the option </span></span></span><span style="color: rgb(204, 0, 0); font-weight: bold;">
--download-mpich=1</span> or <span style="font-weight: bold; color: rgb(204, 0, 0);">--download-mpich=0<br><br><span style="color: rgb(51, 0, 51);"><span style="font-weight: bold;">Regards to All.<br>Amjad Ali.<br></span>
</span></span><br><div><span class="gmail_quote">On 9/4/07, <b class="gmail_sendername">Matthew Knepley</b> <<a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a>> wrote:</span><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
On 9/4/07, amjad ali <<a href="mailto:amjad11@gmail.com">amjad11@gmail.com</a>> wrote:<br>> Hello Satish, hoping you will be fine and happy.<br>><br>> Is it necessary to have MPICH2 already installed (seaprately), before
<br>> installing/configuring PETSc.<br><br>If this is Windows, yes it is necessary (if you want to run in<br>parallel). If not,<br>you can use --with-mpi=0.<br><br>> Do we need at all MPICH2 separately to wprk with PETSc or installing PETSc
<br>> will automatically have MPI implementation within that?<br><br>PETSc can automatically build MPI, but not on Windows unfortunately.<br><br> Thanks,<br><br> Matt<br><br>> regards,<br>> Amjad Ali.<br>>
<br>> Also please answer my Yesterday's question:<br>><br>> What is your opinion about using PETSc for developing/running appliactions<br>> on MS Windows (Compute Cluster server) based cluster?<br>><br>
> Is it better/equal to Linux based system in performance?<br>><br>> regards,<br>> Amjad Ali.<br>><br>><br>><br>> On 8/21/07, amjad ali <<a href="mailto:amjad11@gmail.com">amjad11@gmail.com</a>> wrote:
<br>> ><br>> > Hello Satish,<br>> > Thanks a lot for your kind answers. These are definitely helpful in<br>> clearifying concepts.<br>> ><br>> > Regards,<br>> > Amjad Ali<br>> >
<br>> ><br>> ><br>> > On 8/21/07, Satish Balay <<a href="mailto:petsc-maint@mcs.anl.gov">petsc-maint@mcs.anl.gov</a> > wrote:<br>> > > On Mon, 20 Aug 2007, amjad ali wrote:<br>> > >
<br>> > > > Hello,<br>> > > > Thanks Balay for your kind reply.<br>> > > ><br>> > > > As my system is:<br>> > > > Total cpu cores =4x4 = 16. And there are 2 memory banks per node, so
<br>> total<br>> > > > = 8 memory banks for the complete cluster.<br>> > > ><br>> > > > Let me ask my question in a better way:<br>> > > > *Is this system is suffiently good (giving decent performance) to RUN
<br>> > > > my "PETSc based Parallel CFD/FEM code" on this system *?<br>> > ><br>> > > it should be fine.<br>> > ><br>> > > > Can you please suggest any other *BETTER* PC-cluster for my PETSc
<br>> code?<br>> > ><br>> > > Inte Core Duos appear to have much higher mermory bandwidth - so they<br>> > > can potentially do better [this is unverified/speculation. Also the<br>> > > latencies are higher - so I'm not sure if this hurts performance
<br>> > > much].<br>> > ><br>> > > > > Also - we recommend doing the development on machines where its easy<br>> > > > > to run debuggers [your desktop or laptop etc..] - and not a parallel
<br>> > > > > machine - where such things are hard to do.<br>> > > ><br>> > > > How to install and use PETSc on desktop (single machine)? Is it<br>> straight<br>> > > > forward?
<br>> > ><br>> > > Yes [assuming your desktop is linux] its straight forward. Just use<br>> > > the configure option --download-mpich=1. You'll be able to run<br>> > > parallel codes on this machine [by default MPICH will use multiple
<br>> > > processes for the parallel run, and all these processes will run on<br>> > > your desktop].<br>> > ><br>> > > When developing code - performance is not critical, so the above model
<br>> > > will be more convinent than developing on a cluster.<br>> > ><br>> > > Satish<br>> > ><br>> ><br>> ><br>><br>><br><br><br>--<br>What most experimenters take for granted before they begin their
<br>experiments is infinitely more interesting than any results to which<br>their experiments lead.<br>-- Norbert Wiener<br></blockquote></div><br>