COSMA v2.0.1

This release fixes the issues when building COSMA on some linux distros, with OpenBLAS backend.

Production release of COSMA

This is the first production release of COSMA. It brings a lot of bug-fixes and performance improvements. Some of the most important updates are the following:

• Faster GPU backend:
  • pinning/unpinning of host memory amortized
  • better stream synchronization
  • tiling mechanism improved
• Faster memory access: using huge pages (2M)
• Highly-optimized pxgemm (scalapack) wrapper:
  • layout transformation optimized, using maximum-weighted perfect matching
  • COSMA can use the initial layout directly, if the layout transformation is too expensive in some cases.
• Portability:
  • Hybrid version: ported to both NVIDIA and AMD GPUs.
  • CPU-only version: supports MKL, OpenBLAS, Cray-libsci and custom gemm backends.
• Usability:
  • Trivial integration: to use our code, it is enough to link to the library, without changing the user-code.
  • Spack-installable
• Bug-fixes:
  • correctness tested on up to 1024 nodes of Piz Daint Supercomputer (Cray XC50).

First release of COSMA

First release of COSMA.

Beta release of COSMA

0.1

Fixed a typo.