This site is currently under development for the RSE-ops community.
Published:
Reading time: 1 min
mpibind is a memory-driven algorithm to map parallel hybrid applications to the underlying heterogeneous hardware resources transparently, efficiently, and portably. Unlike other mappings, its primary design point is the memory system, including the cache hierarchy. Compute elements are selected based on a memory mapping and not vice versa. Several new incremental versions have been released simultaneously:
Help improve its content by opening a Pull Request on GitHub.