Expressing Parallelism on Many-Core for Deterministic Discrete Ordinates Transport

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

4 Citations (Scopus)
37 Downloads (Pure)

Abstract

In this paper we demonstrate techniques for increas- ing the node-level parallelism of a deterministic discrete ordinates neutral particle transport algorithm on a structured mesh to exploit many-core technologies. Transport calculations form a large part of the computational workload of physical simulations and so good performance is vital for the simulations to complete in reasonable time. We will demonstrate our approach utilizing the SNAP mini-app, which gives a simplified implementation of the full transport algorithm but remains similar enough to the real algorithm to act as a useful proxy for research purposes.

We present an OpenCL implementation of our improved algorithm which demonstrates a speedup of up to 2.5x the transport sweep performance on a many-core GPGPU device compared to a state-of-the-art multi-core node; the first time this scale of speedup has been achieved for algorithms of this class.
Original languageEnglish
Title of host publication2015 IEEE International Conference on Cluster Computing
Subtitle of host publicationWorkshop on Representative Applications
DOIs
Publication statusPublished - 8 Sep 2015
EventInternational Workshop on Representative Applications (WRAp) - IEEE Cluster, Chicago, United States
Duration: 8 Sep 20158 Sep 2015

Workshop

WorkshopInternational Workshop on Representative Applications (WRAp)
CountryUnited States
CityChicago
Period8/09/158/09/15

Fingerprint Dive into the research topics of 'Expressing Parallelism on Many-Core for Deterministic Discrete Ordinates Transport'. Together they form a unique fingerprint.

Cite this