An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC

Jung Min Park; Jung Wook Park; Cheong Ghil Kim; Gi Ho Park; Shin Dug Kim

An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC

Jung Min Park, Jung Wook Park, Cheong Ghil Kim, Gi Ho Park, Shin Dug Kim

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

Abstract

As the video CODECs become complex increasingly according to the convergence of digital equipments designed for different kinds of specific purposes, the demand of higher performance for video processing is continuously needed. The fast search methods are well known as a representative approach to overcome the performance bottleneck of video encoding. But they require huge computations and dominate most of the running time of video encoding. So, a parallel processing is required to enhance the performance. In general, the fast search methods are difficult to be parallelized because their memory access patterns are not followed by a definite form. This paper proposes a new method for parallelizing a fast search method using linear SIMD architecture with only one-to-one connection between a PE and a memory bank. The simulation results show that the proposed method can reduce nearly 20 percent of running time in the motion estimation (ME).

Original language	English
Title of host publication	20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007
Publisher	International Society for Computers and Their Applications (ISCA)
Pages	169-174
Number of pages	6
ISBN (Electronic)	9781604233926
Publication status	Published - 2007
Event	20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007 - Las Vegas, United States Duration: 2007 Sept 24 → 2007 Sept 26

Publication series

Name	20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007

Conference

Conference	20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007
Country/Territory	United States
City	Las Vegas
Period	07/9/24 → 07/9/26

Bibliographical note

Funding Information:
This work was supported by Samsung Co. “Design of Adaptive Parallel Accelerator and DSP/SIMD Hybrid System for High Performace Media Applications” project.

Publisher Copyright:
© 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. All rights reserved.

All Science Journal Classification (ASJC) codes

Software
Computer Networks and Communications
Hardware and Architecture

Cite this

Park, J. M., Park, J. W., Kim, C. G., Park, G. H., & Kim, S. D. (2007). An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC. In 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007 (pp. 169-174). (20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007). International Society for Computers and Their Applications (ISCA).

Park, Jung Min ; Park, Jung Wook ; Kim, Cheong Ghil et al. / An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC. 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. International Society for Computers and Their Applications (ISCA), 2007. pp. 169-174 (20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007).

@inproceedings{db81d0502e834386807850eaf9a4b76b,

title = "An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC",

abstract = "As the video CODECs become complex increasingly according to the convergence of digital equipments designed for different kinds of specific purposes, the demand of higher performance for video processing is continuously needed. The fast search methods are well known as a representative approach to overcome the performance bottleneck of video encoding. But they require huge computations and dominate most of the running time of video encoding. So, a parallel processing is required to enhance the performance. In general, the fast search methods are difficult to be parallelized because their memory access patterns are not followed by a definite form. This paper proposes a new method for parallelizing a fast search method using linear SIMD architecture with only one-to-one connection between a PE and a memory bank. The simulation results show that the proposed method can reduce nearly 20 percent of running time in the motion estimation (ME).",

author = "Park, {Jung Min} and Park, {Jung Wook} and Kim, {Cheong Ghil} and Park, {Gi Ho} and Kim, {Shin Dug}",

note = "Funding Information: This work was supported by Samsung Co. “Design of Adaptive Parallel Accelerator and DSP/SIMD Hybrid System for High Performace Media Applications” project. Publisher Copyright: {\textcopyright} 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. All rights reserved.; 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007 ; Conference date: 24-09-2007 Through 26-09-2007",

year = "2007",

language = "English",

series = "20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007",

publisher = "International Society for Computers and Their Applications (ISCA)",

pages = "169--174",

booktitle = "20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007",

address = "United States",

}

Park, JM, Park, JW, Kim, CG, Park, GH & Kim, SD 2007, An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC. in 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007, International Society for Computers and Their Applications (ISCA), pp. 169-174, 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007, Las Vegas, United States, 07/9/24.

An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC. / Park, Jung Min; Park, Jung Wook; Kim, Cheong Ghil et al.
20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. International Society for Computers and Their Applications (ISCA), 2007. p. 169-174 (20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007).

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

TY - GEN

T1 - An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC

AU - Park, Jung Min

AU - Park, Jung Wook

AU - Kim, Cheong Ghil

AU - Park, Gi Ho

AU - Kim, Shin Dug

N1 - Funding Information: This work was supported by Samsung Co. “Design of Adaptive Parallel Accelerator and DSP/SIMD Hybrid System for High Performace Media Applications” project. Publisher Copyright: © 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. All rights reserved.

PY - 2007

Y1 - 2007

N2 - As the video CODECs become complex increasingly according to the convergence of digital equipments designed for different kinds of specific purposes, the demand of higher performance for video processing is continuously needed. The fast search methods are well known as a representative approach to overcome the performance bottleneck of video encoding. But they require huge computations and dominate most of the running time of video encoding. So, a parallel processing is required to enhance the performance. In general, the fast search methods are difficult to be parallelized because their memory access patterns are not followed by a definite form. This paper proposes a new method for parallelizing a fast search method using linear SIMD architecture with only one-to-one connection between a PE and a memory bank. The simulation results show that the proposed method can reduce nearly 20 percent of running time in the motion estimation (ME).

AB - As the video CODECs become complex increasingly according to the convergence of digital equipments designed for different kinds of specific purposes, the demand of higher performance for video processing is continuously needed. The fast search methods are well known as a representative approach to overcome the performance bottleneck of video encoding. But they require huge computations and dominate most of the running time of video encoding. So, a parallel processing is required to enhance the performance. In general, the fast search methods are difficult to be parallelized because their memory access patterns are not followed by a definite form. This paper proposes a new method for parallelizing a fast search method using linear SIMD architecture with only one-to-one connection between a PE and a memory bank. The simulation results show that the proposed method can reduce nearly 20 percent of running time in the motion estimation (ME).

UR - http://www.scopus.com/inward/record.url?scp=85132808486&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85132808486&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85132808486

T3 - 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007

SP - 169

EP - 174

BT - 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007

PB - International Society for Computers and Their Applications (ISCA)

T2 - 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007

Y2 - 24 September 2007 through 26 September 2007

ER -

Park JM, Park JW, Kim CG, Park GH, Kim SD. An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC. In 20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007. International Society for Computers and Their Applications (ISCA). 2007. p. 169-174. (20th International Conference on Parallel and Distributed Computing Systems, PDCS 2007).

An Efficient Scheme for Parallelizing Fast Search Algorithm on SIMD Architecture in H.264/AVC

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this