Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
FPGA community forums and blogs on community.intel.com are migrating to the new Altera Community and are read-only. For urgent support needs during this transition, please visit the FPGA Design Resources page or contact an Altera Authorized Distributor.

how to use Cluster PARDISO and ScaLAPACK functions in python ctypes

Chaowen_G_
Beginner
789 Views

Hi:

I know I can use python ctypes to load libmkl_rt.so use mkl. However, libmkl_rt .so does not have all the Cluster PARDISO and ScaLAPACK functions. If I want to use such functions, I needs to -lmkl_scalapack_lp64 -lmkl_intel_lp64 -lmkl_core -lmkl_intel_thread -lmkl_blacs_intelmpi_lp64 -liomp5. But ctypes just can load one shared library, I can not just load mkl_scalapack_lp64 or mkl_intel_lp64, because they all have undefined symbols. I want to ask how to use Cluster PARDISO and ScaLAPACK functions in python ctypes?

0 Kudos
1 Reply
Ying_H_Intel
Moderator
789 Views

Hi Chaowen, 

I guess there is more than one way to link third-party static library or dynamic library to python.

But the first thing in my mind, as cluster Pardiso and scalapack is not in libmkl_rt.so, you can build one  Custom DLL, which link the function you wanted, then link the "custom.dll" instead of libmkl_rt.so. 

You can find the "custom dll " build tool in MKL install folder and some guide in the on-line articles. 

https://software.intel.com/en-us/articles/using-intel-mkl-in-your-python-programs

Best Regards,

Ying 

0 Kudos
Reply