Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf
Online MPI trace compression using event flow graphs and wavelets
KTH, Centra, SeRC - Swedish e-Science Research Centre.ORCID-id: 0000-0001-9693-6265
KTH, Centra, SeRC - Swedish e-Science Research Centre. KTH, Skolan för datavetenskap och kommunikation (CSC), Beräkningsvetenskap och beräkningsteknik (CST).ORCID-id: 0000-0002-9901-9857
2016 (Engelska)Ingår i: Procedia Computer Science, Elsevier, 2016, s. 1497-1506Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Performance analysis of scientific parallel applications is essential to use High Performance Computing (HPC) infrastructures efficiently. Nevertheless, collecting detailed data of large-scale parallel programs and long-running applications is infeasible due to the huge amount of performance information generated. Even though there are no technological constraints in storing Terabytes of performance data, the constant flushing of such data to disk introduces a massive overhead into the application that makes the performance measurements worthless. This paper explores the use of Event flow graphs together with wavelet analysis and EZW-encoding to provide MPI event traces that are orders of magnitude smaller while preserving accurate information on timestamped events. Our mechanism compresses the performance data online while the application runs, thus, reducing the pressure put on the I/O system due to buffer flushing. As a result, we achieve lower application perturbation, reduced performance data output, and the possibility to monitor longer application runs. © The Authors. Published by Elsevier B.V.

Ort, förlag, år, upplaga, sidor
Elsevier, 2016. s. 1497-1506
Nyckelord [en]
Event flow graphs, EZW coding, MPI performance monitoring, Trace compression, Wavelets, Application programs, Graphic methods, Wavelet analysis, Event-flow graph, Performance monitoring, Flow graphs
Nationell ämneskategori
Elektroteknik och elektronik
Identifikatorer
URN: urn:nbn:se:kth:diva-194600DOI: 10.1016/j.procs.2016.05.471Scopus ID: 2-s2.0-84978517854OAI: oai:DiVA.org:kth-194600DiVA, id: diva2:1044118
Konferens
International Conference on Computational Science, ICCS 2016, 6 June 2016 through 8 June 2016
Anmärkning

Conference Paper. QC 20161102

Tillgänglig från: 2016-11-02 Skapad: 2016-10-31 Senast uppdaterad: 2016-11-02Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas i DiVA

Övriga länkar

Förlagets fulltextScopus

Personposter BETA

Aguilar, XavierLaure, Erwin

Sök vidare i DiVA

Av författaren/redaktören
Aguilar, XavierLaure, Erwin
Av organisationen
SeRC - Swedish e-Science Research CentreBeräkningsvetenskap och beräkningsteknik (CST)
Elektroteknik och elektronik

Sök vidare utanför DiVA

GoogleGoogle Scholar

doi
urn-nbn

Altmetricpoäng

doi
urn-nbn
Totalt: 424 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Annat format
Fler format
Språk
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Annat språk
Fler språk
Utmatningsformat
  • html
  • text
  • asciidoc
  • rtf