Mainstream vs. Emerging HPC: Metrics, Trade-Offs and Lessons Learned

Milan Radulovic, Kazi Asifuzzaman, Darko Zivanovic, Nikola Rajovic, Guillaume Colin De Verdiere, Dirk Pleiter, Manolis Marazakisl, Nikolaos Kallimanis, Paul Carpenter, Petar Radojkovic, Eduard Ayguade

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Scopus citations

Abstract

Various servers with different characteristics and architectures are hitting the market, and their evaluation and comparison in terms of HPC features is complex and multidimensional. In this paper, we share our experience of evaluating a diverse set of HPC systems, consisting of three mainstream and five emerging architectures. We evaluate the performance and power efficiency using prominent HPC benchmarks, High-Performance Linpack (HPL) and High Performance Conjugate Gradients (HPCG), and expand our analysis using publicly available specialized kernel benchmarks, targeting specific system components. In addition to a large body of quantitative results, we emphasize six usually overlooked aspects of the HPC platforms evaluation, and share our conclusions and lessons learned. Overall, we believe that this paper will improve the evaluation and comparison of HPC platforms, making a first step towards a more reliable and uniform methodology.

Original languageEnglish
Title of host publicationProceedings - 2018 30th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages250-257
Number of pages8
ISBN (Electronic)9781538677698
DOIs
StatePublished - Jul 2 2018
Externally publishedYes
Event30th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2018 - Lyon, France
Duration: Sep 24 2018Sep 27 2018

Publication series

NameProceedings - 2018 30th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2018

Conference

Conference30th International Symposium on Computer Architecture and High Performance Computing, SBAC-PAD 2018
Country/TerritoryFrance
CityLyon
Period09/24/1809/27/18

Funding

ACKNOWLEDGEMENT This work was supported by the Spanish Ministry of Science and Technology (project TIN2015-65316-P), Generalitat de Catalunya (contracts 2014-SGR-1051 and 2014-SGR-1272), Severo Ochoa Programme (SEV-2015-0493) of the Spanish Government; and the European Union’s Horizon 2020 research and innovation programme under ExaNoDe project (grant agreement No 671578).

FundersFunder number
Severo Ochoa ProgrammeSEV-2015-0493
Horizon 2020 Framework Programme671578, 800898
Generalitat de Catalunya2014-SGR-1272, 2014-SGR-1051
Ministerio de Ciencia y TecnologíaTIN2015-65316-P

    Keywords

    • BytelFLOP ratio
    • HPC
    • emerging and mainstream architectures
    • energy efficiency
    • memory latency and bandwidth
    • weak vs. strong cores

    Fingerprint

    Dive into the research topics of 'Mainstream vs. Emerging HPC: Metrics, Trade-Offs and Lessons Learned'. Together they form a unique fingerprint.

    Cite this