The Open Community Runtime on the Intel Knights Landing Architecture

The Open Community Runtime on the Intel Knights Landing Architecture

Abstract

The Intel Xeon Phi Knights Landing manycore processor comes with new interesting features: on-chip high-bandwidth memory and several user-selectable NUMA configurations. In this paper, we look into how these affect applications that target the Open Community Runtime (OCR), an asynchronous tasked-based runtime system for future parallel architectures. We have extended our OCR runtime to make it NUMA aware and to allow it to use the high-bandwidth memory. We have conducted a range of experiments, comparing OpenMP, TBB, our OCR implementation, and the reference OCR implementation on different machine configurations using a memory intensive seismic simulation.

Grafik Top
Authors
  • Dokulil, Jiri
  • Benkner, Siegfried
  • Yaghob, J.
Grafik Top
Shortfacts
Category
Paper in Conference Proceedings or in Workshop Proceedings (Paper)
Event Title
2nd International Workshop on Ultrascale Computing for Early Researchers (UCER 2017), in conjunction with ICA3PP, Helsinki, Finland, August 21-23, 2017
Divisions
Scientific Computing
Event Location
Helsinki
Event Type
Workshop
Event Dates
21 -23 August, 2017
Date
21 August 2017
Export
Grafik Top