Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks

Luay Alawneh; Abdelwahab Hamou-Lhadj; Jameleddine Hassine

doi:10.1109/SANER.2015.7081821

Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks

Luay Alawneh^*
, Abdelwahab Hamou-Lhadj
, Jameleddine Hassine

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Scopus citations

Abstract

There exist several tools for analyzing traces generated from HPC (High Performance Computing) applications, used by software engineers for debugging and other maintenance tasks. These tools, however, use different formats to represent HPC traces, which hinders interoperability and data exchange. At the present time, there is no standard metamodel that represents HPC trace concepts and their relations. In this paper, we argue that the lack of a common metamodel is a serious impediment for effective analysis for this class of software systems. We aim to fill this void by presenting MTF2 (MPI Trace Format2) - a metamodel for representing HPC system traces. MTF2 is built with expressiveness and scalability in mind. Scalability, an important requirement when working with large traces, is achieved by adopting graph theory concepts to compact large traces. We show through a case study that a trace represented in MTF2 can be in average 49% smaller than a trace represented in a format that does not consider compaction.

Original language	English
Title of host publication	2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	111-120
Number of pages	10
ISBN (Electronic)	9781479984695
DOIs	https://doi.org/10.1109/SANER.2015.7081821
State	Published - 8 Apr 2015

Publication series

Name	2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Keywords

HPC trace analysis
High performance computing systems
inter-process communication traces
metamodeling

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Science Applications
Software

Access to Document

10.1109/SANER.2015.7081821

Cite this

Alawneh, L., Hamou-Lhadj, A., & Hassine, J. (2015). Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks. In 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings (pp. 111-120). Article 7081821 (2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SANER.2015.7081821

Alawneh, Luay ; Hamou-Lhadj, Abdelwahab ; Hassine, Jameleddine. / Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks. 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 111-120 (2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings).

@inproceedings{5c106cb3a6604b3aa7aa21401ef4a7e7,

title = "Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks",

abstract = "There exist several tools for analyzing traces generated from HPC (High Performance Computing) applications, used by software engineers for debugging and other maintenance tasks. These tools, however, use different formats to represent HPC traces, which hinders interoperability and data exchange. At the present time, there is no standard metamodel that represents HPC trace concepts and their relations. In this paper, we argue that the lack of a common metamodel is a serious impediment for effective analysis for this class of software systems. We aim to fill this void by presenting MTF2 (MPI Trace Format2) - a metamodel for representing HPC system traces. MTF2 is built with expressiveness and scalability in mind. Scalability, an important requirement when working with large traces, is achieved by adopting graph theory concepts to compact large traces. We show through a case study that a trace represented in MTF2 can be in average 49\% smaller than a trace represented in a format that does not consider compaction.",

keywords = "HPC trace analysis, High performance computing systems, inter-process communication traces, metamodeling",

author = "Luay Alawneh and Abdelwahab Hamou-Lhadj and Jameleddine Hassine",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.",

year = "2015",

month = apr,

day = "8",

doi = "10.1109/SANER.2015.7081821",

language = "English",

series = "2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "111--120",

booktitle = "2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings",

address = "United States",

}

Alawneh, L, Hamou-Lhadj, A & Hassine, J 2015, Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks. in 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings., 7081821, 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings, Institute of Electrical and Electronics Engineers Inc., pp. 111-120. https://doi.org/10.1109/SANER.2015.7081821

Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks. / Alawneh, Luay; Hamou-Lhadj, Abdelwahab; Hassine, Jameleddine.
2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. p. 111-120 7081821 (2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks

AU - Alawneh, Luay

AU - Hamou-Lhadj, Abdelwahab

AU - Hassine, Jameleddine

PY - 2015/4/8

Y1 - 2015/4/8

N2 - There exist several tools for analyzing traces generated from HPC (High Performance Computing) applications, used by software engineers for debugging and other maintenance tasks. These tools, however, use different formats to represent HPC traces, which hinders interoperability and data exchange. At the present time, there is no standard metamodel that represents HPC trace concepts and their relations. In this paper, we argue that the lack of a common metamodel is a serious impediment for effective analysis for this class of software systems. We aim to fill this void by presenting MTF2 (MPI Trace Format2) - a metamodel for representing HPC system traces. MTF2 is built with expressiveness and scalability in mind. Scalability, an important requirement when working with large traces, is achieved by adopting graph theory concepts to compact large traces. We show through a case study that a trace represented in MTF2 can be in average 49% smaller than a trace represented in a format that does not consider compaction.

AB - There exist several tools for analyzing traces generated from HPC (High Performance Computing) applications, used by software engineers for debugging and other maintenance tasks. These tools, however, use different formats to represent HPC traces, which hinders interoperability and data exchange. At the present time, there is no standard metamodel that represents HPC trace concepts and their relations. In this paper, we argue that the lack of a common metamodel is a serious impediment for effective analysis for this class of software systems. We aim to fill this void by presenting MTF2 (MPI Trace Format2) - a metamodel for representing HPC system traces. MTF2 is built with expressiveness and scalability in mind. Scalability, an important requirement when working with large traces, is achieved by adopting graph theory concepts to compact large traces. We show through a case study that a trace represented in MTF2 can be in average 49% smaller than a trace represented in a format that does not consider compaction.

KW - HPC trace analysis

KW - High performance computing systems

KW - inter-process communication traces

KW - metamodeling

UR - https://www.scopus.com/pages/publications/84928684065

U2 - 10.1109/SANER.2015.7081821

DO - 10.1109/SANER.2015.7081821

M3 - Conference contribution

AN - SCOPUS:84928684065

T3 - 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings

SP - 111

EP - 120

BT - 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Alawneh L, Hamou-Lhadj A, Hassine J. Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks. In 2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. p. 111-120. 7081821. (2015 IEEE 22nd International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Proceedings). doi: 10.1109/SANER.2015.7081821

Towards a common metamodel for traces of high performance computing systems to enable software analysis tasks

Abstract

Publication series

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this