Automatic topology mapping of diverse large-scale parallel applications

Topology-aware mapping aims at assigning tasks to processors in a way that minimizes network load, thus reducing the time spent waiting for communication to complete. Many mapping schemes and algorithms have been proposed. Some are application or domain specific, and others require significant effort by developers or users to successfully apply them. Moreover, a task mapping algorithm by itself is not enough to map the diverse set of applications that exist. Applications can have distinct communication patterns, from point-to-point communication with neighbors in a virtual process grid, to irregular point-to-point communication, to different types of collectives with differing group sizes, and any combination of the above. These patterns should be analyzed, and critical patterns extracted and automatically provided to the mapping algorithm, all without specialized user input. To our knowledge, this problem has not been addressed before for the general case. In this paper, we propose a complete and automatic mapping system that does not require special user involvement, works with any application, and whose mapping performs better than existing schemes, for a wide range of communication patterns and machine topologies. This makes it suitable for online mapping of HPC applications in many different scenarios. We evaluate our scheme with several applications exhibiting different communication patterns (including collectives) on machines with 3D torus, 5D torus and fat-tree network topologies, and show up to 2.2x performance improvements.