Parceive: Interactive parallelization based on dynamic analysis

Andreas Wilhelm; Bharatkumar Sharmay; Ranajoy Malakary; Tobias Schule; Michael Gerndt

doi:10.1109/PCODA.2015.7067176

Parceive: Interactive parallelization based on dynamic analysis

Andreas Wilhelm
, Bharatkumar Sharmay
, Ranajoy Malakary
, Tobias Schule
, Michael Gerndt

Informatics 10 - Associate Professorship of Architecture of Parallel and Distributed Systems

Technical University of Munich
Siemens AG

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

7 Scopus citations

Abstract

The advent of multicore processors poses an urgent need for tools to parallelize legacy software. Automatic parallelization methods are usually limited to the instruction level or to simple loops. However, identifying parallelism in industrial applications additionally requires deep program comprehension. To solve this problem, we propose Parceive, an interactive tool that aids parallelization of software at various granularity levels. Parceive uses dynamic binary instrumentation to trace programs written in C/C++. The collected data dependencies and profiling information are then analyzed to visualize potential parallelization candidates. This approach helps developers to comprehend the application and to parallelize their software. In this paper, we motivate our approach, illustrate the architecture of Parceive, and highlight optimizations to cope with industrial applications. A case study shows the usefulness of our approach.

Original language	English
Title of host publication	2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1-6
Number of pages	6
ISBN (Electronic)	9781467369176
DOIs	https://doi.org/10.1109/PCODA.2015.7067176
State	Published - 24 Mar 2015
Event	6th International Workshop on Program Comprehension through Dynamic Analysis, PCODA 2015, co-located with the 22nd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 - Montreal, Canada Duration: 2 Mar 2015 → 2 Mar 2015

Publication series

Name	2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings
Volume	2015-March

Conference

Conference	6th International Workshop on Program Comprehension through Dynamic Analysis, PCODA 2015, co-located with the 22nd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015
Country/Territory	Canada
City	Montreal
Period	2/03/15 → 2/03/15

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Access to Document

10.1109/PCODA.2015.7067176

Cite this

Wilhelm, A., Sharmay, B., Malakary, R., Schule, T., & Gerndt, M. (2015). Parceive: Interactive parallelization based on dynamic analysis. In 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings (pp. 1-6). Article 7067176 (2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings; Vol. 2015-March). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PCODA.2015.7067176

Wilhelm, Andreas ; Sharmay, Bharatkumar ; Malakary, Ranajoy et al. / Parceive : Interactive parallelization based on dynamic analysis. 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 1-6 (2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings).

@inproceedings{4ef5c1135ee14ac1950d5160433b3751,

title = "Parceive: Interactive parallelization based on dynamic analysis",

abstract = "The advent of multicore processors poses an urgent need for tools to parallelize legacy software. Automatic parallelization methods are usually limited to the instruction level or to simple loops. However, identifying parallelism in industrial applications additionally requires deep program comprehension. To solve this problem, we propose Parceive, an interactive tool that aids parallelization of software at various granularity levels. Parceive uses dynamic binary instrumentation to trace programs written in C/C++. The collected data dependencies and profiling information are then analyzed to visualize potential parallelization candidates. This approach helps developers to comprehend the application and to parallelize their software. In this paper, we motivate our approach, illustrate the architecture of Parceive, and highlight optimizations to cope with industrial applications. A case study shows the usefulness of our approach.",

author = "Andreas Wilhelm and Bharatkumar Sharmay and Ranajoy Malakary and Tobias Schule and Michael Gerndt",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 6th International Workshop on Program Comprehension through Dynamic Analysis, PCODA 2015, co-located with the 22nd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015 ; Conference date: 02-03-2015 Through 02-03-2015",

year = "2015",

month = mar,

day = "24",

doi = "10.1109/PCODA.2015.7067176",

language = "English",

series = "2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings",

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

pages = "1--6",

booktitle = "2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings",

}

Wilhelm, A, Sharmay, B, Malakary, R, Schule, T & Gerndt, M 2015, Parceive: Interactive parallelization based on dynamic analysis. in 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings., 7067176, 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings, vol. 2015-March, Institute of Electrical and Electronics Engineers Inc., pp. 1-6, 6th International Workshop on Program Comprehension through Dynamic Analysis, PCODA 2015, co-located with the 22nd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015, Montreal, Canada, 2/03/15. https://doi.org/10.1109/PCODA.2015.7067176

Parceive: Interactive parallelization based on dynamic analysis. / Wilhelm, Andreas; Sharmay, Bharatkumar; Malakary, Ranajoy et al.
2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2015. p. 1-6 7067176 (2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings; Vol. 2015-March).

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

TY - GEN

T1 - Parceive

T2 - 6th International Workshop on Program Comprehension through Dynamic Analysis, PCODA 2015, co-located with the 22nd IEEE International Conference on Software Analysis, Evolution, and Reengineering, SANER 2015

AU - Wilhelm, Andreas

AU - Sharmay, Bharatkumar

AU - Malakary, Ranajoy

AU - Schule, Tobias

AU - Gerndt, Michael

PY - 2015/3/24

Y1 - 2015/3/24

N2 - The advent of multicore processors poses an urgent need for tools to parallelize legacy software. Automatic parallelization methods are usually limited to the instruction level or to simple loops. However, identifying parallelism in industrial applications additionally requires deep program comprehension. To solve this problem, we propose Parceive, an interactive tool that aids parallelization of software at various granularity levels. Parceive uses dynamic binary instrumentation to trace programs written in C/C++. The collected data dependencies and profiling information are then analyzed to visualize potential parallelization candidates. This approach helps developers to comprehend the application and to parallelize their software. In this paper, we motivate our approach, illustrate the architecture of Parceive, and highlight optimizations to cope with industrial applications. A case study shows the usefulness of our approach.

AB - The advent of multicore processors poses an urgent need for tools to parallelize legacy software. Automatic parallelization methods are usually limited to the instruction level or to simple loops. However, identifying parallelism in industrial applications additionally requires deep program comprehension. To solve this problem, we propose Parceive, an interactive tool that aids parallelization of software at various granularity levels. Parceive uses dynamic binary instrumentation to trace programs written in C/C++. The collected data dependencies and profiling information are then analyzed to visualize potential parallelization candidates. This approach helps developers to comprehend the application and to parallelize their software. In this paper, we motivate our approach, illustrate the architecture of Parceive, and highlight optimizations to cope with industrial applications. A case study shows the usefulness of our approach.

UR - https://www.scopus.com/pages/publications/84990935049

U2 - 10.1109/PCODA.2015.7067176

DO - 10.1109/PCODA.2015.7067176

M3 - Conference contribution

AN - SCOPUS:84990935049

T3 - 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings

SP - 1

EP - 6

BT - 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 2 March 2015 through 2 March 2015

ER -

Wilhelm A, Sharmay B, Malakary R, Schule T, Gerndt M. Parceive: Interactive parallelization based on dynamic analysis. In 2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings. Institute of Electrical and Electronics Engineers Inc. 2015. p. 1-6. 7067176. (2015 IEEE 6th International Workshop on Program Comprehension Through Dynamic Analysis, PCODA 2015 - Proceedings). doi: 10.1109/PCODA.2015.7067176

Parceive: Interactive parallelization based on dynamic analysis

Abstract

Publication series

Conference

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this