A joint hybrid training event of German and Austrian National Competence Centres for High-Performance Computing, SIDE and EuroCC-Austria, Hybrid Programming in HPC – MPI+X, will be organized by the HLRS in cooperation with the ASC Research Center, TU Wien, and NHR@FAU from 10 to 12 February 2026.
Most HPC systems are clusters of shared memory nodes, and many use accelerators, e.g., GPUs. To use such systems efficiently, both memory consumption and communication time must be optimized. Therefore, hybrid programming may combine the distributed memory parallelization on the node interconnect with the shared memory parallelization inside each node.
This course examines the strengths and weaknesses of various parallel programming models on clusters of SMP nodes, both with and without accelerators. Special focus is given to multi-socket, multi-core systems operating in highly parallel environments, regardless of the presence of accelerators. Additionally, the course reviews the shared memory programming interface introduced in MPI-3.0, which can be integrated with inter-node MPI communication.
Through numerous case studies and micro-benchmarks, students will explore performance aspects of hybrid programming. Hands-on sessions are included throughout the course, and a dedicated how-to section presents tools for hybrid programming, such as thread and process placement support and performance analysis.
