Performance model for Master/Worker hybrid applications on multicore clusters

Abel Castellanos, Andreu Moreno, Joan Sorribes, Tomas Margalef

Research output: Chapter in BookChapterResearchpeer-review

2 Citations (Scopus)

Abstract

There are several parallel applications that are implemented using a Master/Worker parallel/distributed programming paradigm. Applications using this predefined programming structure can be easily implemented using message passing programming libraries (MPI). Moreover, the multicore features present nowadays on CPU architecture can be exploited at the node level by applying thread parallelism (OpenMP). In order to exploit the benefits of both two levels of parallelism, the Master/Worker applications can be implemented as hybrid applications. However, reaching the expected performance indexes is not trivial because there are several parameters (number of nodes, number of threads per node, thread affinity and data distribution among all nodes) that must be tuned for each particular application or even during its execution to reach a successful performance. On the other hand, the application workload may change drastically during successive executions, so those parameters need to be modified according to this behavior. Additionally, cache memory architecture in multicore systems directly influences the performance and this behaviour must be deeply analysed. In this paper we present a proposal to model the performance of hybrid Master/Worker applications on multicore systems considering the issues outlined above. In particular, this model determines dynamically at runtime the configuration of the appropriate number of workers and threads of the hybrid application to achieve the best possible performance.

Original languageEnglish
Title of host publication2013 IEEE 10th International Confernce on High Performance Computing and Communications (HPCC 2013)
Place of Publication(US)
PublisherIEEE Computer Society
Pages201-217
Number of pages16
Edition1
ISBN (Print)978-1-4799-0973-5
DOIs
Publication statusPublished - 2014

Publication series

NameProceedings - 2013 IEEE International Conference on High Performance Computing and Communications, HPCC 2013 and 2013 IEEE International Conference on Embedded and Ubiquitous Computing, EUC 2013

Keywords

  • Hybrid applications
  • Master/Worker
  • MPI
  • multicore
  • OpenMP
  • Performance model

Fingerprint

Dive into the research topics of 'Performance model for Master/Worker hybrid applications on multicore clusters'. Together they form a unique fingerprint.

Cite this