Owenson, Andrew M B, Wright, Steven A., Bunt, Richard A., Jarvis, Stephen A., Ho, Y. K. and Street, Matthew (2018) Developing and using a geometric multigrid, unstructured grid mini-application to assess many-core architectures. In: 26th Euromicro International Conference on Parallel, Distributed and Network-based Processing, Cambridge, UK, 21-23 Mar 2018 (Submitted)

Preview

PDF WRAP-developing-geometric-multigrid-unstructured-assess-Owenson-2018.pdf - Submitted Version - Requires a PDF viewer. Download (821Kb) | Preview

Request Changes to record.

Abstract

Achieving high-performance of large scientific codes is a difficult task. This has led to the development of numerous mini-applications that are more tractable to analyse, while retaining performance characteristics of their full-sized counterparts. These “mini-apps” also enable faster hardware evaluation, and for sensitive codes allow evaluation of systems outside of access approval processes.

In this paper we develop a mini-application of a geometric multigrid, unstructured grid Computational Fluid Dynamics (CFD) code, designed to exhibit similar performance characteristics without sharing code. We detail our experiences developing this application, using guidelines detailed in existing research, and contribute further additions to these to aid future mini-application developers. Our application is validated against the inviscid flux routine of HYDRA, a CFD code developed by Rolls-Royce, which confirms that the parent kernel and mini-application share fundamental causes of parallel inefficiency.

We then use the mini-application to assess the impact of Intel's Knights Landing (KNL) on performance. We find that the mini-app and parent kernel continue to share scaling characteristics, however a comparison with Broadwell performance exposed significant differences between the kernels that were undetected by the validation.

Item Type:	Conference Item (Paper)
Subjects:	Q Science > QA Mathematics > QA76 Electronic computers. Computer science. Computer software T Technology > TA Engineering (General). Civil engineering (General)
Divisions:	Faculty of Science > Computer Science
Library of Congress Subject Headings (LCSH):	Computational fluid dynamics -- Software, High performance computing
Official Date:	2018
Dates:	Date Event 2018 Published 11 December 2017 Accepted 10 November 2017 Submitted 12 January 2018 Completion
Status:	Peer Reviewed
Publication Status:	Submitted
Access rights to Published version:	Restricted or Subscription Access
Funder:	Engineering and Physical Sciences Research Council (EPSRC), Intel Corporation
RIOXX Funder/Project Grant:	Project/Grant ID RIOXX Funder Name Funder ID Clean Sky project Rolls-Royce http://dx.doi.org/10.13039/501100000767 UNSPECIFIED Engineering and Physical Sciences Research Council http://dx.doi.org/10.13039/501100000266 CASE award 15220082 Intel Corporation http://dx.doi.org/10.13039/100002418
Conference Paper Type:	Paper
Title of Event:	26th Euromicro International Conference on Parallel, Distributed and Network-based Processing
Type of Event:	Conference
Location of Event:	Cambridge, UK
Date(s) of Event:	21-23 Mar 2018
Related URLs:	Organisation

Request changes or add full text files to a record

Repository staff actions (login required)

View Item

Downloads