Ändra sökning
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • 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
MPI Trace Compression Using Event Flow Graphs
KTH, Skolan för datavetenskap och kommunikation (CSC), High Performance Computing and Visualization (HPCViz).ORCID-id: 0000-0001-9693-6265
Ludwig-Maximilians-Universitat (LMU). (Computer Science Department, MNM Team)
KTH, Skolan för datavetenskap och kommunikation (CSC), High Performance Computing and Visualization (HPCViz).ORCID-id: 0000-0002-9901-9857
2014 (Engelska)Konferensbidrag, Publicerat paper (Refereegranskat)
Abstract [en]

Understanding how parallel applications behave is crucial for using high-performance computing (HPC) resources efficiently. However, the task of performance analysis is becoming increasingly difficult due to the growing complexity of scientific codes and the size of machines. Even though many tools have been developed over the past years to help in this task, current approaches either only offer an overview of the application discarding temporal information, or they generate huge trace files that are often difficult to handle.

In this paper we propose the use of event flow graphs for monitoring MPI applications, a new and different approach that balances the low overhead of profiling tools with the abundance of information available from tracers. Event flow graphs are captured with very low overhead, require orders of magnitude less storage than standard trace files, and can still recover the full sequence of events in the application. We test this new approach with the NERSC-8/Trinity Benchmark suite and achieve compression ratios up to 119x.

Ort, förlag, år, upplaga, sidor
2014. 1-12 s.
Serie
Lecture Notes in Computer Science, ISSN 0302-9743 ; 8632
Nyckelord [en]
MPI event flow graphs, trace compression, trace reconstruction, performance monitoring
Nationell ämneskategori
Datorsystem
Forskningsämne
Datalogi
Identifikatorer
URN: urn:nbn:se:kth:diva-165042Scopus ID: 2-s2.0-84958532986OAI: oai:DiVA.org:kth-165042DiVA: diva2:806807
Konferens
Euro-Par 2014 Parallel Processing
Anmärkning

QC 20150423. QC 20160314

Tillgänglig från: 2015-04-21 Skapad: 2015-04-21 Senast uppdaterad: 2017-04-28Bibliografiskt granskad
Ingår i avhandling
1. Towards Scalable Performance Analysis of MPI Parallel Applications
Öppna denna publikation i ny flik eller fönster >>Towards Scalable Performance Analysis of MPI Parallel Applications
2015 (Engelska)Licentiatavhandling, sammanläggning (Övrigt vetenskapligt)
Abstract [en]

  A considerably fraction of science discovery is nowadays relying on computer simulations. High Performance Computing  (HPC) provides scientists with the means to simulate processes ranging from climate modeling to protein folding. However, achieving good application performance and making an optimal use of HPC resources is a heroic task due to the complexity of parallel software. Therefore, performance tools  and runtime systems that help users to execute  applications in the most optimal way are of utmost importance in the landscape of HPC.  In this thesis, we explore different techniques to tackle the challenges of collecting, storing, and using  fine-grained performance data. First, we investigate the automatic use of real-time performance data in order to run applications in an optimal way. To that end, we present a prototype of an adaptive task-based runtime system that uses real-time performance data for task scheduling. This runtime system has a performance monitoring component that provides real-time access to the performance behavior of anapplication while it runs. The implementation of this monitoring component is presented and evaluated within this thesis. Secondly, we explore lossless compression approaches  for MPI monitoring. One of the main problems that  performance tools face is the huge amount of fine-grained data that can be generated from an instrumented application. Collecting fine-grained data from a program is the best method to uncover the root causes of performance bottlenecks, however, it is unfeasible with extremely parallel applications  or applications with long execution times. On the other hand, collecting coarse-grained data is scalable but  sometimes not enough to discern the root cause of a performance problem. Thus, we propose a new method for performance monitoring of MPI programs using event flow graphs. Event flow graphs  provide very low overhead in terms of execution time and  storage size, and can be used to reconstruct fine-grained trace files of application events ordered in time.

Ort, förlag, år, upplaga, sidor
Stockholm: KTH Royal Institute of Technology, 2015. viii, 39 s.
Serie
TRITA-CSC-A, ISSN 1653-5723 ; 2015:05
Nyckelord
parallel computing, performance monitoring, performance tools, event flow graphs
Nationell ämneskategori
Datorsystem
Forskningsämne
Datalogi
Identifikatorer
urn:nbn:se:kth:diva-165043 (URN)978-91-7595-518-6 (ISBN)
Presentation
2015-05-20, The Visualization Studio, room 4451, Lindstedtsvägen 5, KTH, Stockholm, 10:00 (Engelska)
Opponent
Handledare
Anmärkning

QC 20150508

Tillgänglig från: 2015-05-08 Skapad: 2015-04-21 Senast uppdaterad: 2015-05-08Bibliografiskt granskad

Open Access i DiVA

Fulltext saknas

Scopus

Sök vidare i DiVA

Av författaren/redaktören
Aguilar, XavierLaure, Erwin
Av organisationen
High Performance Computing and Visualization (HPCViz)
Datorsystem

Sök vidare utanför DiVA

GoogleGoogle Scholar

urn-nbn

Altmetricpoäng

urn-nbn
Totalt: 153 träffar
RefereraExporteraLänk till posten
Permanent länk

Direktlänk
Referera
Referensformat
  • apa
  • 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