The OPS Domain Specific Abstraction for Multi-Block Structured Grid Computations

Istvan Reguly, Gihan Mudalige, Michael Giles, Daniel Curran, Simon N McIntosh-Smith

Research output: Contribution to conferenceConference Paperpeer-review

Abstract

Code maintainability, performance portability and future proofing are some of the key challenges in this era of rapid change in High Performance Computing. Domain Specific Languages address these challenges by focusing on a single application domain and providing a high-level programming approach, and then subsequently using domain knowledge to deliver high performance on various hardware.

In this paper, we introduce the OPS high-level abstraction and framework aimed at multi-block structured grid computations, and discuss some of its key design points; we demonstrate how the OPS DSL can be embedded in C/C++, and the API made to look like a traditional library, and how through a combination of simple text manipulation and back-end logic we can enable execution on a diverse range of hardware using different parallel programming approaches.

Relying on the access-execute description of the OPS abstraction, we introduce a number of automated execution techniques that enable distributed memory parallelization, optimization of communication patterns, checkpointing and cache-blocking. Using performance results from CloverLeaf from the Mantevo suite of applications, we demonstrate the utility of OPS.
Original languageEnglish
Number of pages10
Publication statusAccepted/In press - 17 Nov 2014
EventSC14 - The Ernest N. Morial Convention Center, New Orleans, United States
Duration: 16 Nov 201421 Nov 2014
http://sc14.supercomputing.org/

Conference

ConferenceSC14
CountryUnited States
CityNew Orleans
Period16/11/1421/11/14
OtherSC Conference is dedicated to showcasing work in high performance computing, networking, storage and analysis by the international HPC community.
Internet address

Fingerprint

Dive into the research topics of 'The OPS Domain Specific Abstraction for Multi-Block Structured Grid Computations'. Together they form a unique fingerprint.

Cite this