Bringing task and data parallelism to analysis of climate model output

Robert Jacob, Jayesh Krishna, Xiabing Xu, Sheri Mickelson, Tim Tautges, Mike Wilde, Robert Latham, Ian Foster, Robert Ross, Mark Hereld, Jay Larson, Pavel Bochev, Kara Peterson, Mark Taylor, Karen Schuchardt, Jain Yin, Don Middleton, Mary Haley, David Brown, Wei HuangDennis Shea, Richard Brownrigg, Mariana Vertenstein, Kwan-Liu Ma, Jingrong Xie

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

Abstract

Climate models are both outputting larger and larger amounts of data and are doing it on more sophisticated numerical grids. The tools climate scientists have used to analyze climate output, an essential component of climate modeling, are single threaded and assume rectangular structured grids in their analysis algorithms. We are bringing both task- and data-parallelism to the analysis of climate model output. We have created a new data-parallel library, the Parallel Gridded Analysis Library (ParGAL) which can read in data using parallel I/O, store the data on a compete representation of the structured or unstructured mesh and perform sophisticated analysis on the data in parallel. ParGAL has been used to create a parallel version of a script-based analysis and visualization package. Finally, we have also taken current workflows and employed task-based parallelism to decrease the total execution time.

Original languageEnglish (US)
Title of host publicationProceedings - 2012 SC Companion
Subtitle of host publicationHigh Performance Computing, Networking Storage and Analysis, SCC 2012
Pages1493-1495
Number of pages3
DOIs
StatePublished - Dec 1 2012
Event2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012 - Salt Lake City, UT, United States
Duration: Nov 10 2012Nov 16 2012

Other

Other2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012
CountryUnited States
CitySalt Lake City, UT
Period11/10/1211/16/12

Fingerprint

Climate models
Visualization

Keywords

  • climate
  • parallel analysis
  • unstructured grids

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Software

Cite this

Jacob, R., Krishna, J., Xu, X., Mickelson, S., Tautges, T., Wilde, M., ... Xie, J. (2012). Bringing task and data parallelism to analysis of climate model output. In Proceedings - 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012 (pp. 1493-1495). [6496065] https://doi.org/10.1109/SC.Companion.2012.282

Bringing task and data parallelism to analysis of climate model output. / Jacob, Robert; Krishna, Jayesh; Xu, Xiabing; Mickelson, Sheri; Tautges, Tim; Wilde, Mike; Latham, Robert; Foster, Ian; Ross, Robert; Hereld, Mark; Larson, Jay; Bochev, Pavel; Peterson, Kara; Taylor, Mark; Schuchardt, Karen; Yin, Jain; Middleton, Don; Haley, Mary; Brown, David; Huang, Wei; Shea, Dennis; Brownrigg, Richard; Vertenstein, Mariana; Ma, Kwan-Liu; Xie, Jingrong.

Proceedings - 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012. 2012. p. 1493-1495 6496065.

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

Jacob, R, Krishna, J, Xu, X, Mickelson, S, Tautges, T, Wilde, M, Latham, R, Foster, I, Ross, R, Hereld, M, Larson, J, Bochev, P, Peterson, K, Taylor, M, Schuchardt, K, Yin, J, Middleton, D, Haley, M, Brown, D, Huang, W, Shea, D, Brownrigg, R, Vertenstein, M, Ma, K-L & Xie, J 2012, Bringing task and data parallelism to analysis of climate model output. in Proceedings - 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012., 6496065, pp. 1493-1495, 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012, Salt Lake City, UT, United States, 11/10/12. https://doi.org/10.1109/SC.Companion.2012.282
Jacob R, Krishna J, Xu X, Mickelson S, Tautges T, Wilde M et al. Bringing task and data parallelism to analysis of climate model output. In Proceedings - 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012. 2012. p. 1493-1495. 6496065 https://doi.org/10.1109/SC.Companion.2012.282
Jacob, Robert ; Krishna, Jayesh ; Xu, Xiabing ; Mickelson, Sheri ; Tautges, Tim ; Wilde, Mike ; Latham, Robert ; Foster, Ian ; Ross, Robert ; Hereld, Mark ; Larson, Jay ; Bochev, Pavel ; Peterson, Kara ; Taylor, Mark ; Schuchardt, Karen ; Yin, Jain ; Middleton, Don ; Haley, Mary ; Brown, David ; Huang, Wei ; Shea, Dennis ; Brownrigg, Richard ; Vertenstein, Mariana ; Ma, Kwan-Liu ; Xie, Jingrong. / Bringing task and data parallelism to analysis of climate model output. Proceedings - 2012 SC Companion: High Performance Computing, Networking Storage and Analysis, SCC 2012. 2012. pp. 1493-1495
@inproceedings{4bbfdb2704c44a278072aa9d8ecf3f18,
title = "Bringing task and data parallelism to analysis of climate model output",
abstract = "Climate models are both outputting larger and larger amounts of data and are doing it on more sophisticated numerical grids. The tools climate scientists have used to analyze climate output, an essential component of climate modeling, are single threaded and assume rectangular structured grids in their analysis algorithms. We are bringing both task- and data-parallelism to the analysis of climate model output. We have created a new data-parallel library, the Parallel Gridded Analysis Library (ParGAL) which can read in data using parallel I/O, store the data on a compete representation of the structured or unstructured mesh and perform sophisticated analysis on the data in parallel. ParGAL has been used to create a parallel version of a script-based analysis and visualization package. Finally, we have also taken current workflows and employed task-based parallelism to decrease the total execution time.",
keywords = "climate, parallel analysis, unstructured grids",
author = "Robert Jacob and Jayesh Krishna and Xiabing Xu and Sheri Mickelson and Tim Tautges and Mike Wilde and Robert Latham and Ian Foster and Robert Ross and Mark Hereld and Jay Larson and Pavel Bochev and Kara Peterson and Mark Taylor and Karen Schuchardt and Jain Yin and Don Middleton and Mary Haley and David Brown and Wei Huang and Dennis Shea and Richard Brownrigg and Mariana Vertenstein and Kwan-Liu Ma and Jingrong Xie",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/SC.Companion.2012.282",
language = "English (US)",
isbn = "9780769549569",
pages = "1493--1495",
booktitle = "Proceedings - 2012 SC Companion",

}

TY - GEN

T1 - Bringing task and data parallelism to analysis of climate model output

AU - Jacob, Robert

AU - Krishna, Jayesh

AU - Xu, Xiabing

AU - Mickelson, Sheri

AU - Tautges, Tim

AU - Wilde, Mike

AU - Latham, Robert

AU - Foster, Ian

AU - Ross, Robert

AU - Hereld, Mark

AU - Larson, Jay

AU - Bochev, Pavel

AU - Peterson, Kara

AU - Taylor, Mark

AU - Schuchardt, Karen

AU - Yin, Jain

AU - Middleton, Don

AU - Haley, Mary

AU - Brown, David

AU - Huang, Wei

AU - Shea, Dennis

AU - Brownrigg, Richard

AU - Vertenstein, Mariana

AU - Ma, Kwan-Liu

AU - Xie, Jingrong

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Climate models are both outputting larger and larger amounts of data and are doing it on more sophisticated numerical grids. The tools climate scientists have used to analyze climate output, an essential component of climate modeling, are single threaded and assume rectangular structured grids in their analysis algorithms. We are bringing both task- and data-parallelism to the analysis of climate model output. We have created a new data-parallel library, the Parallel Gridded Analysis Library (ParGAL) which can read in data using parallel I/O, store the data on a compete representation of the structured or unstructured mesh and perform sophisticated analysis on the data in parallel. ParGAL has been used to create a parallel version of a script-based analysis and visualization package. Finally, we have also taken current workflows and employed task-based parallelism to decrease the total execution time.

AB - Climate models are both outputting larger and larger amounts of data and are doing it on more sophisticated numerical grids. The tools climate scientists have used to analyze climate output, an essential component of climate modeling, are single threaded and assume rectangular structured grids in their analysis algorithms. We are bringing both task- and data-parallelism to the analysis of climate model output. We have created a new data-parallel library, the Parallel Gridded Analysis Library (ParGAL) which can read in data using parallel I/O, store the data on a compete representation of the structured or unstructured mesh and perform sophisticated analysis on the data in parallel. ParGAL has been used to create a parallel version of a script-based analysis and visualization package. Finally, we have also taken current workflows and employed task-based parallelism to decrease the total execution time.

KW - climate

KW - parallel analysis

KW - unstructured grids

UR - http://www.scopus.com/inward/record.url?scp=84876582441&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84876582441&partnerID=8YFLogxK

U2 - 10.1109/SC.Companion.2012.282

DO - 10.1109/SC.Companion.2012.282

M3 - Conference contribution

AN - SCOPUS:84876582441

SN - 9780769549569

SP - 1493

EP - 1495

BT - Proceedings - 2012 SC Companion

ER -