2-3-4 combination for parallel compression on the K computer

Chongke Bi, Kenji Ono

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

The development of supercomputers has successfully helped us to carry on complicated simulation with exploded size of dataset. For visualizing such kind of large-scale dataset, reducing the data size by using compression methods is one of the most useful approach. Moreover, parallelization of compression algorithm can greatly improve the efficiency and resolve the limitation of memory size. However, in parallel compression algorithm, interprocessor communication is indispensable, while it is also a bottleneck problem, especially for the general cases that the number of processors is not power-of-Two. Parallel POD (proper orthogonal decomposition) compression algorithm is such an example, the number of time steps must be power-of-Two for the binary swap scheme. A method that can fully resolve this problem with low computational cost will be very popular. In this paper, we proposed such an approach called 2-3-4 combination approach, which can be simply implemented and also reach high performance of parallel computing algorithms. Furthermore, our method can obtain the best balance among all parallel computing processors. This is achieved by transferring the non-power-of-Two problem into power-of-Two problem to fully use the best balance feature of binary swap method. We evaluate our approach through applying it to the parallel POD compression algorithm on the K computer.

Original languageEnglish
Title of host publicationProceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014
PublisherIEEE Computer Society
Pages281-285
Number of pages5
ISBN (Print)9781479928736
DOIs
Publication statusPublished - Jan 1 2014
Event2014 7th IEEE Pacific Visualization Symposium, PacificVis 2014 - Yokohama, Kanagawa, Japan
Duration: Mar 4 2014Mar 7 2014

Publication series

NameIEEE Pacific Visualization Symposium
ISSN (Print)2165-8765
ISSN (Electronic)2165-8773

Other

Other2014 7th IEEE Pacific Visualization Symposium, PacificVis 2014
CountryJapan
CityYokohama, Kanagawa
Period3/4/143/7/14

Fingerprint

Parallel processing systems
Decomposition
Supercomputers
Data storage equipment
Communication
Costs

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition
  • Hardware and Architecture
  • Software

Cite this

Bi, C., & Ono, K. (2014). 2-3-4 combination for parallel compression on the K computer. In Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014 (pp. 281-285). [6787182] (IEEE Pacific Visualization Symposium). IEEE Computer Society. https://doi.org/10.1109/PacificVis.2014.28

2-3-4 combination for parallel compression on the K computer. / Bi, Chongke; Ono, Kenji.

Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014. IEEE Computer Society, 2014. p. 281-285 6787182 (IEEE Pacific Visualization Symposium).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Bi, C & Ono, K 2014, 2-3-4 combination for parallel compression on the K computer. in Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014., 6787182, IEEE Pacific Visualization Symposium, IEEE Computer Society, pp. 281-285, 2014 7th IEEE Pacific Visualization Symposium, PacificVis 2014, Yokohama, Kanagawa, Japan, 3/4/14. https://doi.org/10.1109/PacificVis.2014.28
Bi C, Ono K. 2-3-4 combination for parallel compression on the K computer. In Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014. IEEE Computer Society. 2014. p. 281-285. 6787182. (IEEE Pacific Visualization Symposium). https://doi.org/10.1109/PacificVis.2014.28
Bi, Chongke ; Ono, Kenji. / 2-3-4 combination for parallel compression on the K computer. Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014. IEEE Computer Society, 2014. pp. 281-285 (IEEE Pacific Visualization Symposium).
@inproceedings{cd44daeafbcd405ba14c47bf91d39fa1,
title = "2-3-4 combination for parallel compression on the K computer",
abstract = "The development of supercomputers has successfully helped us to carry on complicated simulation with exploded size of dataset. For visualizing such kind of large-scale dataset, reducing the data size by using compression methods is one of the most useful approach. Moreover, parallelization of compression algorithm can greatly improve the efficiency and resolve the limitation of memory size. However, in parallel compression algorithm, interprocessor communication is indispensable, while it is also a bottleneck problem, especially for the general cases that the number of processors is not power-of-Two. Parallel POD (proper orthogonal decomposition) compression algorithm is such an example, the number of time steps must be power-of-Two for the binary swap scheme. A method that can fully resolve this problem with low computational cost will be very popular. In this paper, we proposed such an approach called 2-3-4 combination approach, which can be simply implemented and also reach high performance of parallel computing algorithms. Furthermore, our method can obtain the best balance among all parallel computing processors. This is achieved by transferring the non-power-of-Two problem into power-of-Two problem to fully use the best balance feature of binary swap method. We evaluate our approach through applying it to the parallel POD compression algorithm on the K computer.",
author = "Chongke Bi and Kenji Ono",
year = "2014",
month = "1",
day = "1",
doi = "10.1109/PacificVis.2014.28",
language = "English",
isbn = "9781479928736",
series = "IEEE Pacific Visualization Symposium",
publisher = "IEEE Computer Society",
pages = "281--285",
booktitle = "Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014",
address = "United States",

}

TY - GEN

T1 - 2-3-4 combination for parallel compression on the K computer

AU - Bi, Chongke

AU - Ono, Kenji

PY - 2014/1/1

Y1 - 2014/1/1

N2 - The development of supercomputers has successfully helped us to carry on complicated simulation with exploded size of dataset. For visualizing such kind of large-scale dataset, reducing the data size by using compression methods is one of the most useful approach. Moreover, parallelization of compression algorithm can greatly improve the efficiency and resolve the limitation of memory size. However, in parallel compression algorithm, interprocessor communication is indispensable, while it is also a bottleneck problem, especially for the general cases that the number of processors is not power-of-Two. Parallel POD (proper orthogonal decomposition) compression algorithm is such an example, the number of time steps must be power-of-Two for the binary swap scheme. A method that can fully resolve this problem with low computational cost will be very popular. In this paper, we proposed such an approach called 2-3-4 combination approach, which can be simply implemented and also reach high performance of parallel computing algorithms. Furthermore, our method can obtain the best balance among all parallel computing processors. This is achieved by transferring the non-power-of-Two problem into power-of-Two problem to fully use the best balance feature of binary swap method. We evaluate our approach through applying it to the parallel POD compression algorithm on the K computer.

AB - The development of supercomputers has successfully helped us to carry on complicated simulation with exploded size of dataset. For visualizing such kind of large-scale dataset, reducing the data size by using compression methods is one of the most useful approach. Moreover, parallelization of compression algorithm can greatly improve the efficiency and resolve the limitation of memory size. However, in parallel compression algorithm, interprocessor communication is indispensable, while it is also a bottleneck problem, especially for the general cases that the number of processors is not power-of-Two. Parallel POD (proper orthogonal decomposition) compression algorithm is such an example, the number of time steps must be power-of-Two for the binary swap scheme. A method that can fully resolve this problem with low computational cost will be very popular. In this paper, we proposed such an approach called 2-3-4 combination approach, which can be simply implemented and also reach high performance of parallel computing algorithms. Furthermore, our method can obtain the best balance among all parallel computing processors. This is achieved by transferring the non-power-of-Two problem into power-of-Two problem to fully use the best balance feature of binary swap method. We evaluate our approach through applying it to the parallel POD compression algorithm on the K computer.

UR - http://www.scopus.com/inward/record.url?scp=84899576727&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84899576727&partnerID=8YFLogxK

U2 - 10.1109/PacificVis.2014.28

DO - 10.1109/PacificVis.2014.28

M3 - Conference contribution

AN - SCOPUS:84899576727

SN - 9781479928736

T3 - IEEE Pacific Visualization Symposium

SP - 281

EP - 285

BT - Proceedings - 2014 IEEE Pacific Visualization Symposium, PacificVis 2014

PB - IEEE Computer Society

ER -