Cholesky Decomposition using ScaLAPACK and Intel's Fortran Compiler

Cholesky Decomposition using ScaLAPACK and Intel's Fortran Compiler

Hi, I am trying to implement an parallel Cholesky decomposition in a fortran code that I am writing. I have seen that the Intel fortran compiler has ScaLAPACK in its libraries and was wondering if anyone out there has used this, and if they have any examples of doing the Cholesky factorization for a positive definite matrix in parallel using Fortran. Any help would be great, as I am just starting out with even learning parallel programming. Thanks.

3 posts / 0 new
Last post
For more complete information about compiler optimizations, see our Optimization Notice.

ScaLAPACK is actually part of the Intel Math Kernel Library, not the Intel Fortran Compiler. I am moving your thread over to the MKL forum for response.

Steve - Intel Developer Support

Well, it depends on what you are trying to do here. If you need a parallel implementation of Cholesky decomposition, you can simply call the LAPACK function in MKL, DPOTRF. If, on the other hand you want to understand writing the code for Cholesky decomposition and try to parallelize that, I would recommend either Numerical Recipes or going to www.netlib.org and get the LAPACK source code and understand it.

As I am writing this I realize you may be interested in the parallelization of this code for a cluster. The ScaLAPACK code, at the same netlib site, might be of help there.

Bruce

Leave a Comment

Please sign in to add a comment. Not a member? Join today