0

I'm trying to solve A*x = b where A has complex values and is dense.

I used cusolverDnCgeqrf() method from cuSolverDN library to do the QR decomposition for one linear set of equations. However, I want to do this several times to speed up the processing.

Is there a "batched" version of this method? Or is there another CUDA library I can use?

David K
  • 124
  • 4
  • 2
    "Is there a "batched" version of this method?" Have you looked at the cusolver [documentation](http://docs.nvidia.com/cuda/cusolver/index.html#cuds-function-reference) ? "Or is there another CUDA library I can use?" From [here](https://stackoverflow.com/help/on-topic): "Questions asking us to recommend or find a book, tool, software library, tutorial or other off-site resource are off-topic for Stack Overflow" – Robert Crovella Jul 14 '17 at 00:16
  • This might help: https://forums.developer.nvidia.com/t/openmp-cublas-for-batched-matrix-solves-using-lu/127265 – InfiniteElementMethod Jun 12 '20 at 21:46

1 Answers1

2

You can use Magma batched QR: http://icl.cs.utk.edu/projectsfiles/magma/doxygen/group__group__qr__batched.html#details

Or Nvidia batched library: https://devblogs.nvidia.com/parallelforall/parallel-direct-solvers-with-cusolver-batched-qr/

I am not sure if there are python wrappers for them yet. I want to add that batched version of many solvers are currently available, either through Magma or Nvidia.

There is not a single standard yet, but it is underway, it is discussed in batched blas workshops: here

http://www.netlib.org/utk/people/JackDongarra/WEB-PAGES/Batched-BLAS-2017/ and here:

http://www.netlib.org/utk/people/JackDongarra/WEB-PAGES/Batched-BLAS-2016/

The draft is ready and I hope there would be a standard Batched BLAS soon.

Aznaveh
  • 558
  • 8
  • 27