Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications

W. Caarls; P.P. Jonker; H. Corporaal

doi:10.1109/IPDPS.2006.1639351

Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications

W. Caarls
, P.P. Jonker
, H. Corporaal

Electronic Systems

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

13 Citations (Scopus)

Abstract

Algorithmic skeletons can be used to write architecture independent programs, shielding application developers from the details of a parallel implementation. In this paper, we present a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects. By using skeletons to control the iteration of kernel functions, we provide a stream programming language that is better tailored to the user as well as the underlying architecture. Skeleton merging allows us to reduce the overheads usually associated with breaking an application into small kernels. We have implemented an example image processing application on a heterogeneous embedded prototype platform consisting of an SIMD and ILP processor, and show that a significant speedup can be achieved without requiring knowledge of data parallel processing.

Original language	English
Title of host publication	20th International Parallel and Distributed Processing Symposium, IPDPS 2006
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	9
ISBN (Print)	1-4244-0054-6
DOIs	https://doi.org/10.1109/IPDPS.2006.1639351
Publication status	Published - 2006
Event	20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006 - Rhodes Island, Greece Duration: 25 Apr 2006 → 29 Apr 2006

Conference

Conference	20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006
Country/Territory	Greece
City	Rhodes Island
Period	25/04/06 → 29/04/06

Access to Document

10.1109/IPDPS.2006.1639351

Cite this

@inproceedings{c4e84a246ba5410ea8060070eeb49d08,

title = "Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications",

abstract = "Algorithmic skeletons can be used to write architecture independent programs, shielding application developers from the details of a parallel implementation. In this paper, we present a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects. By using skeletons to control the iteration of kernel functions, we provide a stream programming language that is better tailored to the user as well as the underlying architecture. Skeleton merging allows us to reduce the overheads usually associated with breaking an application into small kernels. We have implemented an example image processing application on a heterogeneous embedded prototype platform consisting of an SIMD and ILP processor, and show that a significant speedup can be achieved without requiring knowledge of data parallel processing.",

author = "W. Caarls and P.P. Jonker and H. Corporaal",

year = "2006",

doi = "10.1109/IPDPS.2006.1639351",

language = "English",

isbn = "1-4244-0054-6",

booktitle = "20th International Parallel and Distributed Processing Symposium, IPDPS 2006",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006 ; Conference date: 25-04-2006 Through 29-04-2006",

}

Caarls, W, Jonker, PP & Corporaal, H 2006, Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications. in 20th International Parallel and Distributed Processing Symposium, IPDPS 2006., 1639351, Institute of Electrical and Electronics Engineers, Piscataway, 20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006, Rhodes Island, Greece, 25/04/06. https://doi.org/10.1109/IPDPS.2006.1639351

Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications. / Caarls, W.; Jonker, P.P.; Corporaal, H.
20th International Parallel and Distributed Processing Symposium, IPDPS 2006. Piscataway: Institute of Electrical and Electronics Engineers, 2006. 1639351.

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

TY - GEN

T1 - Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications

AU - Caarls, W.

AU - Jonker, P.P.

AU - Corporaal, H.

PY - 2006

Y1 - 2006

N2 - Algorithmic skeletons can be used to write architecture independent programs, shielding application developers from the details of a parallel implementation. In this paper, we present a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects. By using skeletons to control the iteration of kernel functions, we provide a stream programming language that is better tailored to the user as well as the underlying architecture. Skeleton merging allows us to reduce the overheads usually associated with breaking an application into small kernels. We have implemented an example image processing application on a heterogeneous embedded prototype platform consisting of an SIMD and ILP processor, and show that a significant speedup can be achieved without requiring knowledge of data parallel processing.

AB - Algorithmic skeletons can be used to write architecture independent programs, shielding application developers from the details of a parallel implementation. In this paper, we present a C-like skeleton implementation language, PEPCI, that uses term rewriting and partial evaluation to specify skeletons for parallel C dialects. By using skeletons to control the iteration of kernel functions, we provide a stream programming language that is better tailored to the user as well as the underlying architecture. Skeleton merging allows us to reduce the overheads usually associated with breaking an application into small kernels. We have implemented an example image processing application on a heterogeneous embedded prototype platform consisting of an SIMD and ILP processor, and show that a significant speedup can be achieved without requiring knowledge of data parallel processing.

UR - https://www.scopus.com/pages/publications/33847140637

U2 - 10.1109/IPDPS.2006.1639351

DO - 10.1109/IPDPS.2006.1639351

M3 - Conference contribution

SN - 1-4244-0054-6

BT - 20th International Parallel and Distributed Processing Symposium, IPDPS 2006

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 20th IEEE International Parallel and Distributed Processing Symposium, IPDPS 2006

Y2 - 25 April 2006 through 29 April 2006

ER -

Algorithmic skeletons for stream programming in embedded heterogeneous parallel image processing applications

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this