<  Back to the Polytechnique Montréal portal

A flexible data-driven approach for execution trace filtering

Kadjo Gwandy Kouamé, Naser Ezzati-Jivan, Michel Dagenais

Paper (2015)

Open Access document in PolyPublie
[img]
Preview
Open Access to the full text of this document
Accepted Version
Terms of Use: All rights reserved
Download (420kB)
Show abstract
Hide abstract

Abstract

Execution traces are frequently used to study systemrun-time behaviour and to detect problems. However, the hugeamount of data in an execution trace may complexify its analysis.Moreover, users are not usually interested in all events of atrace, hence the need for a proper filtering approach. Filteringis used to generate an enhanced trace, with a reduced size andcomplexity, that is easier to analyse. The approach described inthis paper allows to define custom filtering patterns, declarativelyin XML, to concentrate the analysis on the most important andinteresting events. The filtering scenarios include syntaxes todescribe various analysis patterns using finite state machines.The patterns range from very simple event filtering to complexmulti-level event abstraction, covering various types of syntheticbehaviours that can be captured from execution trace data. Thepaper provides the details on this data-driven filtering approachand some interesting use cases for the trace events generated bythe LTTng Linux kernel tracer.

Uncontrolled Keywords

LTTng, Trace Analysis, Performance analysis, Data Filtering

Subjects: 2700 Information technology > 2705 Software and development
2700 Information technology > 2720 Computer systems software
Department: Department of Computer Engineering and Software Engineering
Funders: CRSNG/NSERC
Grant number: CRDPJ468687-14
PolyPublie URL: https://publications.polymtl.ca/2985/
Conference Title: IEEE International Congress on Big Data (BigData Congress 2015)
Conference Location: New York, NY, USA
Conference Date(s): 2015-06-27 - 2015-07-02
Publisher: IEEE
DOI: 10.1109/bigdatacongress.2015.112
Official URL: https://doi.org/10.1109/bigdatacongress.2015.112
Date Deposited: 13 Feb 2018 10:39
Last Modified: 22 Nov 2022 08:06
Cite in APA 7: Kouamé, K. G., Ezzati-Jivan, N., & Dagenais, M. (2015, June). A flexible data-driven approach for execution trace filtering [Paper]. IEEE International Congress on Big Data (BigData Congress 2015), New York, NY, USA (6 pages). https://doi.org/10.1109/bigdatacongress.2015.112

Statistics

Total downloads

Downloads per month in the last year

Origin of downloads

Dimensions

Repository Staff Only

View Item View Item