Dear all,
I have a new APPLE with Intel Core 2 Duo Processor which shows very poor performance with all ALPS Applications requiring linear algebra like fulldiag and dmrg. I have already tried to install the ATLAS and LAPACK routines, which had no effect.
When installing ALPS and ALPS applications I get the following error message:
i686-apple-darwin8-g++-4.0.1: unrecognized option '-pthread'
It seems that the pthreads are not working with the standard c- compiler on my system.
Is it possible that this causes the poor performance? And if so, what can I do to overcome this problem?
Many thanks,
Dirk
Do you configure ALPS using the --with-compiler=macos option?
Matthias Troyer
On 11 Jun 2007, at 15:36, Dirk Schuricht wrote:
Dear all,
I have a new APPLE with Intel Core 2 Duo Processor which shows very poor performance with all ALPS Applications requiring linear algebra like fulldiag and dmrg. I have already tried to install the ATLAS and LAPACK routines, which had no effect.
When installing ALPS and ALPS applications I get the following error message:
i686-apple-darwin8-g++-4.0.1: unrecognized option '-pthread'
It seems that the pthreads are not working with the standard c- compiler on my system.
Is it possible that this causes the poor performance? And if so, what can I do to overcome this problem?
Many thanks,
Dirk
Dear Prof. Troyer,
thank you for your answer. Now, I have configured ALPS using the - with-compiler=macos option. The error message has disappeared, however, the low performance of the linear algebra related routines (in particular dmrg) still remains. The brand-new APPLE is still four to five times slower than my two-years old laptop. The recompilation of ATLAS and LAPACK on the APPLE had no effect on this issue.
Do you have any idea what else I could try?
Many thanks and with kind regards,
Dirk Schuricht
On 11 Jun 2007, at 19:38, Matthias Troyer wrote:
Do you configure ALPS using the --with-compiler=macos option?
Matthias Troyer
On 11 Jun 2007, at 15:36, Dirk Schuricht wrote:
Dear all,
I have a new APPLE with Intel Core 2 Duo Processor which shows very poor performance with all ALPS Applications requiring linear algebra like fulldiag and dmrg. I have already tried to install the ATLAS and LAPACK routines, which had no effect.
When installing ALPS and ALPS applications I get the following error message:
i686-apple-darwin8-g++-4.0.1: unrecognized option '-pthread'
It seems that the pthreads are not working with the standard c- compiler on my system.
Is it possible that this causes the poor performance? And if so, what can I do to overcome this problem?
Many thanks,
Dirk
Your old laptop is an Apple or another machine?
On Jun 12, 2007, at 3:35 PM, Dirk Schuricht wrote:
Dear Prof. Troyer,
thank you for your answer. Now, I have configured ALPS using the - with-compiler=macos option. The error message has disappeared, however, the low performance of the linear algebra related routines (in particular dmrg) still remains. The brand-new APPLE is still four to five times slower than my two-years old laptop. The recompilation of ATLAS and LAPACK on the APPLE had no effect on this issue.
Do you have any idea what else I could try?
Many thanks and with kind regards,
Dirk Schuricht
On 11 Jun 2007, at 19:38, Matthias Troyer wrote:
Do you configure ALPS using the --with-compiler=macos option?
Matthias Troyer
On 11 Jun 2007, at 15:36, Dirk Schuricht wrote:
Dear all,
I have a new APPLE with Intel Core 2 Duo Processor which shows very poor performance with all ALPS Applications requiring linear algebra like fulldiag and dmrg. I have already tried to install the ATLAS and LAPACK routines, which had no effect.
When installing ALPS and ALPS applications I get the following error message:
i686-apple-darwin8-g++-4.0.1: unrecognized option '-pthread'
It seems that the pthreads are not working with the standard c- compiler on my system.
Is it possible that this causes the poor performance? And if so, what can I do to overcome this problem?
Many thanks,
Dirk
Matthias Troyer wrote:
On Jun 12, 2007, at 3:35 PM, Dirk Schuricht wrote:
The recompilation of ATLAS and LAPACK on the APPLE had no effect on this issue.
Recompiling it might have no effect at all if ALPS still chooses Apple's version instead.
My wild guess on the subject: is the code using many small BLAS/LAPACK calls? If so, are you using threading? (if you don't know then you probably are, it's enabled by default).
My experience with vecLib was that it was highly inefficient for small calls as the overhead for threading seems to be huge. Recompiling ATLAS for OSX and disabling threads or using Intel's MKL and exporting OMP_NUM_THREADS=1 helped.
If you download CHUD from the OSX developer's page you can use the 'shark' application for a quick guess on where you're spending your time. If it's threading then you'll see that there.
Regards, Emanuel
comp-phys-alps-users@lists.phys.ethz.ch