Details
-
Improvement
-
Status: Open
-
Major
-
Resolution: Unresolved
-
3.1.0
-
None
-
None
Description
BLAS are highly optimized routines by the industry. If take advantage of it, we can leverage hardware vendors' low-level optimization when available.
Sparse BLAS also has native optimization , but current BLAS in Spark only has native optimization for Dense and naive Java implementation for Sparse.
We would like to introduce native optimization support for Sparse BLAS operations related to machine learning and produce benchmarks.