The LAPACK forum has moved to https://github.com/Reference-LAPACK/lapack/discussions.

Scalapack Matrix Redistribution

Open discussion regarding features, bugs, issues, vendors, etc.

Scalapack Matrix Redistribution

Postby robAtGT » Thu Jul 22, 2010 10:36 am

Is there any built-in functionality for redistributing a matrix to a different blocking factor configuration? It appears that p?lacpy and p?copy only copy to matrices with the exact same blocking factor.

Specifically, I am loading my data in a head node and need to distribute it to a general distributed matrix. I am working on my own function to do this using ?gesd2d and ?gerv2d, but it is neither pretty nor efficient so I would appreciate any advice.

The only example I have found is pdgeread.f http://www.netlib.org/scalapack/examples/pdlaread.f, but it appears to (mis)use p?geadd as a redistribution function by setting alpha = 1 and beta = 0. Is this the simplest, way?

Thanks
robAtGT
 
Posts: 1
Joined: Fri Jul 02, 2010 11:20 pm
Location: Georgia Institute of Technology, Atlanta, GA

Return to User Discussion

Who is online

Users browsing this forum: No registered users and 7 guests