Dynamic code mapping for limited local memory systems

Seung Chul Jung; Aviral Shrivastava; Ke Bai

doi:10.1109/ASAP.2010.5540773

Dynamic code mapping for limited local memory systems

Seung Chul Jung, Aviral Shrivastava, Ke Bai

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

30 Scopus citations

Abstract

This paper presents heuristics for dynamic management of application code on limited local memories present in high-performance multi-core processors. Previous techniques formulate the problem using call graphs, which do not capture the temporal ordering of functions. In addition, they only use a conservative estimate of the interference cost between functions to obtain a mapping. As a result previous techniques are unable to achieve efficient code mapping. Techniques proposed in this paper overcome both these limitations and achieve superior code mapping. Experimental results from executing benchmarks from MiBench onto the Cell processor in the Sony Playstation 3 demonstrate upto 29% and average 12% performance improvement, at tolerable compile-time overhead.

Original language	English (US)
Title of host publication	ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings
Pages	13-20
Number of pages	8
DOIs	https://doi.org/10.1109/ASAP.2010.5540773
State	Published - 2010
Event	21st IEEE International Conference on Application-specific Systems, Architectures and Processors, ASAP 2010 - Rennes, France Duration: Jul 7 2010 → Jul 9 2010

Publication series

Name	Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors
ISSN (Print)	1063-6862

Other

Other	21st IEEE International Conference on Application-specific Systems, Architectures and Processors, ASAP 2010
Country/Territory	France
City	Rennes
Period	7/7/10 → 7/9/10

ASJC Scopus subject areas

Hardware and Architecture
Computer Networks and Communications

Access to Document

10.1109/ASAP.2010.5540773

Cite this

Jung, S. C., Shrivastava, A., & Bai, K. (2010). Dynamic code mapping for limited local memory systems. In ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings (pp. 13-20). Article 5540773 (Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors). https://doi.org/10.1109/ASAP.2010.5540773

Dynamic code mapping for limited local memory systems. / Jung, Seung Chul; Shrivastava, Aviral; Bai, Ke.
ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings. 2010. p. 13-20 5540773 (Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors).

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

Jung, SC, Shrivastava, A & Bai, K 2010, Dynamic code mapping for limited local memory systems. in ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings., 5540773, Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors, pp. 13-20, 21st IEEE International Conference on Application-specific Systems, Architectures and Processors, ASAP 2010, Rennes, France, 7/7/10. https://doi.org/10.1109/ASAP.2010.5540773

Jung SC, Shrivastava A, Bai K. Dynamic code mapping for limited local memory systems. In ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings. 2010. p. 13-20. 5540773. (Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors). doi: 10.1109/ASAP.2010.5540773

@inproceedings{bcfdab2f79d8408e9e9aba25727b5abd,

title = "Dynamic code mapping for limited local memory systems",

abstract = "This paper presents heuristics for dynamic management of application code on limited local memories present in high-performance multi-core processors. Previous techniques formulate the problem using call graphs, which do not capture the temporal ordering of functions. In addition, they only use a conservative estimate of the interference cost between functions to obtain a mapping. As a result previous techniques are unable to achieve efficient code mapping. Techniques proposed in this paper overcome both these limitations and achieve superior code mapping. Experimental results from executing benchmarks from MiBench onto the Cell processor in the Sony Playstation 3 demonstrate upto 29% and average 12% performance improvement, at tolerable compile-time overhead.",

author = "Jung, {Seung Chul} and Aviral Shrivastava and Ke Bai",

year = "2010",

doi = "10.1109/ASAP.2010.5540773",

language = "English (US)",

isbn = "9781424469673",

series = "Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors",

pages = "13--20",

booktitle = "ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings",

note = "21st IEEE International Conference on Application-specific Systems, Architectures and Processors, ASAP 2010 ; Conference date: 07-07-2010 Through 09-07-2010",

}

TY - GEN

T1 - Dynamic code mapping for limited local memory systems

AU - Jung, Seung Chul

AU - Shrivastava, Aviral

AU - Bai, Ke

PY - 2010

Y1 - 2010

N2 - This paper presents heuristics for dynamic management of application code on limited local memories present in high-performance multi-core processors. Previous techniques formulate the problem using call graphs, which do not capture the temporal ordering of functions. In addition, they only use a conservative estimate of the interference cost between functions to obtain a mapping. As a result previous techniques are unable to achieve efficient code mapping. Techniques proposed in this paper overcome both these limitations and achieve superior code mapping. Experimental results from executing benchmarks from MiBench onto the Cell processor in the Sony Playstation 3 demonstrate upto 29% and average 12% performance improvement, at tolerable compile-time overhead.

AB - This paper presents heuristics for dynamic management of application code on limited local memories present in high-performance multi-core processors. Previous techniques formulate the problem using call graphs, which do not capture the temporal ordering of functions. In addition, they only use a conservative estimate of the interference cost between functions to obtain a mapping. As a result previous techniques are unable to achieve efficient code mapping. Techniques proposed in this paper overcome both these limitations and achieve superior code mapping. Experimental results from executing benchmarks from MiBench onto the Cell processor in the Sony Playstation 3 demonstrate upto 29% and average 12% performance improvement, at tolerable compile-time overhead.

UR - http://www.scopus.com/inward/record.url?scp=77955865785&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955865785&partnerID=8YFLogxK

U2 - 10.1109/ASAP.2010.5540773

DO - 10.1109/ASAP.2010.5540773

M3 - Conference contribution

AN - SCOPUS:77955865785

SN - 9781424469673

T3 - Proceedings of the International Conference on Application-Specific Systems, Architectures and Processors

SP - 13

EP - 20

BT - ASAP 10 - 21st IEEE International Conference on Application-Specific Systems, Architectures and Processors, Conference Proceedings

T2 - 21st IEEE International Conference on Application-specific Systems, Architectures and Processors, ASAP 2010

Y2 - 7 July 2010 through 9 July 2010

ER -

Dynamic code mapping for limited local memory systems

Abstract

Publication series

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this