IXPUG Workshop: Many-Core Computing on Intel Processors: Applications, Performance and Best-Practice Solutions

摘要

IXPUG has extended its scope to cover all aspects of HPC systems, and this is reflected in the agenda of this year's IXPUG Workshop. The selected presentations covered traditional HPC workloads and their optimization on Intel Xeon and Xeon Phi platforms as well as experience reports of applying machine learning methods. An update in the area of programming models were given with the presentation of the new features in OpenMP 5. Approaches to achieve portable performance were outlined in the two invited talks.