A new data compression technique for event based program traces

Research output: Contribution to book/conference proceedings/anthology/reportConference contributionContributedpeer-review


The paper presents an innovative solution to the problem of the very huge data sets that are regularly produced by performance tracing techniques-especially on HPC programs. It designs an adapted data compression scheme that takes advantage of regularities frequently found in program traces. Algorithms to reveal repetition patterns in a programs call structure and run time behavior are discussed in detail, solutions to some problems arising on practical application are addressed as well. Two examples demonstrate the capabilities of the approach and document its behavior. Finally, some thoughts are given regarding how the patterns revealed in the process of data compression may assist the automatic analysis of traces.


Original languageEnglish
Title of host publicationComputational Science
EditorsPMA Sloot, D Abramson, AV Bogdanov, JJ Dongarra, AY Zomaya, YE Gorbachev
PublisherSpringer, Berlin [u. a.]
Number of pages10
ISBN (print)3-540-40196-2
Publication statusPublished - 2003

Publication series

SeriesLecture Notes in Computer Science, Volume 2659


TitleInternational Conference on Computational Science 2003
Abbreviated titleICCS 2003
Conference number3
Duration2 - 4 June 2003

External IDs

Scopus 35248887492


Library keywords