Showing results for

- Intel Community
- Software Development SDKs and Libraries
- Intel® oneAPI Math Kernel Library & Intel® Math Kernel Library
- Sparse Blas with extended precision

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

Highlighted
##

Hi,

as I'm trying to calculate eigenvalues with ARPACK of a very big matrix, I am in the need of efficient sparse matrix-vector multiplication routines. Unfortunately, at the same time I need more precision than just double precision. Now my question: Is there any support of the MKL sparse Blas matrix-vector multiplication routines (in particular mkl_*bsrgemv) for complex(16) matrices and vectors or some kind of workaround for mkl_zbsrgemv to gain more precision?

Thanks,

Martin

boeseskimchi

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-10-2011
06:24 AM

10 Views

Sparse Blas with extended precision

as I'm trying to calculate eigenvalues with ARPACK of a very big matrix, I am in the need of efficient sparse matrix-vector multiplication routines. Unfortunately, at the same time I need more precision than just double precision. Now my question: Is there any support of the MKL sparse Blas matrix-vector multiplication routines (in particular mkl_*bsrgemv) for complex(16) matrices and vectors or some kind of workaround for mkl_zbsrgemv to gain more precision?

Thanks,

Martin

2 Replies

Highlighted
##

In complex(16) compilation, public source code would do as well as could be done by detailed hand coding.

TimP

Black Belt

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-10-2011
06:50 AM

10 Views

Highlighted
##

Gennady_F_Intel

Moderator

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

01-10-2011
11:11 AM

10 Views

yes,mkl doesn't support quad precision data types.

For more complete information about compiler optimizations, see our Optimization Notice.