How to parallelize many small SVD's on a GPU

8 visualizaciones (últimos 30 días)
Tim
Tim el 8 de Dic. de 2021
Comentada: Tim el 18 de Dic. de 2021
I'm computing SVD's of millions of small (e.g. 5 x 10) matrices. It is surprisingly fast as a for-loop on the CPU, however I need the computation to be even faster and would like to know if parallelization can be leveraged on a GPU. SVD doesn't currently appear to be a function supported by pagefun for gpuArray objects. I'm using R2020a and the Parallel Computing toolbox. Thanks

Respuesta aceptada

Christine Tobler
Christine Tobler el 9 de Dic. de 2021
As of R2021b, there is a function pagesvd which is supported on both CPU and distributed arrays. It's not supported for gpuArray as of now.
  7 comentarios
Joss Knight
Joss Knight el 18 de Dic. de 2021
gpuArray/pagesvd coming next year, watch this space...
Tim
Tim el 18 de Dic. de 2021
Great news, thank you Joss

Iniciar sesión para comentar.

Más respuestas (1)

Matt J
Matt J el 9 de Dic. de 2021
Editada: Matt J el 9 de Dic. de 2021
SVD doesn't currently appear to be a function supported by pagefun for gpuArray objects
Perhaps not for gpuArrays, but it is supported for distributed arrays:

Categorías

Más información sobre Linear Algebra en Help Center y File Exchange.

Productos


Versión

R2020a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by