You are in : Home » Results & submissions » Select UOA » 11 - Computer Science and Informatics » View submission: University of Edinburgh » Outputs » Detail

Output details

11 - Computer Science and Informatics

University of Edinburgh

Return to search Previous output Next output

Output 315 of 401 in the submission

Output title

Semi-automatic extraction and exploitation of hierarchical pipeline parallelism using profiling information

Type

E - Conference contribution

DOI

10.1145/1854273.1854321

Name of conference/published proceedings

Proceedings of the 19th international conference on Parallel Architectures and Compilation Techniques (PACT '10)

Volume number

Issue number

First page of article

377

ISSN of proceedings

Year of publication

2010

URL

http://dl.acm.org/citation.cfm?doid=1854273.1854321

Number of additional authors

Additional information

<08> Originality: Presents the first methodology to extract pipeline parallelism from legacy sequential applications. Profiling information replaces conservative static analysis. Pipeline stages are automatically identified and balanced and if safe and profitable replicated, functions split, and data parallel loops parallelised. User only checks final program.

Significance: Pipeline parallelism is at the heart of many streaming, multimedia, cryptographic and compression applications. The parallelisation tools from this research have potential to enhance performance of many applications on current and future multicores.

Rigour: Ideas implemented in commercial CoSy software framework, evaluated against large real-world applications. PACT'10 conference had just 17% acceptance rate.

Interdisciplinary

Cross-referral requested

Research group

C - Institute for Computing Systems Architecture

Citation count

Proposed double-weighted

Double-weighted statement

Reserve for a double-weighted output

Non-English

English abstract