A considerable amount of research and engineering went into designing proxy applications, which represent common high-performance computing workloads, to co-design and evaluate the current generation of supercomputers, e.g., RIKEN's Supercomputer Fugaku, ANL's Aurora, and ORNL's Frontier. These scaled-down versions of important scientific applications helped researchers around the world to identify bottlenecks, performance characteristics, and fine-tune architectures for their needs. We present some of our findings which will also influence the next generation of HPC systems. However, not all aspects of these proxy applications are compelling, and to accommodate an extremely-heterogeneous future, we have to reconsider how to co-design supercomputers during the next decade. Hence, we will introduce a new kind of benchmarking set, called Octopodes, which hopefully solves some of the open issues, and can be used to close the ever widening gap between sustained and peak performance of our supercomputers.
Jens is a Research Scientist of the High Performance Big Data Research Team at the RIKEN Center for Computational Science (R-CCS), Japan. He received his PhD from the TU Dresden, Germany, in 2017 for his work on HPC routing algorithms and interconnects. Jens contributed the DFSSSP and Nue routing algorithms to the subnet manager of InfiniBand, and built the first large-scale HyperX prototype at the Tokyo Institute of Technology. His research interests include interconnect networks, performance evaluation and extrapolation, and optimization of parallel applications and architectures.