Tools for Monitoring CPU Usage and Affinity in Multicore Supercomputers

Abstract

Performance boosts in HPC nodes have come from making SIMD units wider and aggressively packing more and more cores in each processor. With multiple processors and so many cores it has become necessary to understand and manage process and thread affinity and pinning. However, affinity tools have not been designed specifically for HPC users to quickly evaluate process affinity and execution location. To fill in the gap, three HPC user-friendly tools, core_usage, show_affinity, and amask, have been designed to eliminate barriers that frustrate users and impede users from evaluating and analyzing affinity for applications. These tools focus on providing convenient methods, easy-to-understand affinity representations for large process counts, process locality, and run-time core load with socket aggregation. These tools will significantly help HPC users, developers and site administrators easily monitor processor utilization from an affinity perspective.