公开(公告)号：US20240095081A1

公开(公告)日：2024-03-21

申请号：US17948159

申请日：2022-09-19

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： Jan Maximilian Mäder ,  Christian Simmendinger ,  Tobias Walter Wolfgang Schiffmann ,  Torsten Wilde

IPC: G06F9/50 ,  G06F11/34

CPC classification number: G06F9/505 ,  G06F9/5044 ,  G06F11/3428 ,  G06F2201/80 ,  G06F2209/5019 ,  G06F2209/505

Abstract: A method for optimizing operations of high-performance computing (HPC) systems includes collecting data associated with a plurality of workload performance profiling counters associated with a workload during runtime of the workload in an HPC system. Based on the collected data, the method includes using a machine-learning technique to classify the workload by determining a workload-specific fingerprint for the workload. The method includes identifying an optimization metric to optimize during running of the workload in the HPC system. The method includes determining an optimal setting for a plurality of tunable hardware execution parameters as measured against the optimization metric by varying at least a portion of the plurality of tunable hardware execution parameters. The method includes storing the workload-specific fingerprint, the optimization metric, and the optimal setting for the plurality of tunable hardware execution parameters as measured against the optimization metric in an architecture-specific knowledge database.