First released in September 2018, Aluminum provides a generic interface to high-performance communication libraries with a focus on allreduce algorithms. Blocking and non-blocking algorithms and GPU-aware algorithms are supported. Aluminum also contains custom implementations of select algorithms to optimize for certain situations.

Improvements included in this release:

• Host-transfer implementations of standard collectives in the MPI-CUDA backend
• Experimental RMA Put/Get operations
• Improved algorithm specification, point-to-point operations, testing, and benchmarks

Learn more:

• Aluminum v0.2 release notes
• Aluminum GitHub repo