Improving MPI independent write performance using a two-stage write-behind buffering method

Wei Keng Liao, Avery Ching, Kenin Coloma, Alok Choudhary, Mahmut Kandemir

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

5 Citations (Scopus)

Abstract

Many large-scale production applications often have very long executions times and require periodic data check-points in order to save the state of the computation for program restart and/or tracing application progress. These write-only operations often dominate the overall application runtime, which makes them a good optimization target. Existing approaches for write-behind data buffering at the MPI I/O level have been proposed, but challenges still exist for addressing system-level I/O issues. We propose a two-stage write-behind buffering scheme for handing checkpoint operations. The first-stage of buffering accumulates write data for better network utilization and the second-stage of buffering enables the alignment for the write requests to the file stripe boundaries. Aligned I/O requests avoid file lock contention that can seriously degrade I/O performance. We present our performance evaluation using BTIO benchmarks on both GPFS and Lustre file systems. With the two-stage buffering, the performance of BTIO through MPI independent I/O is significantly improved and even surpasses that of collective I/O.

Original languageEnglish (US)
Title of host publicationProceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM
DOIs
StatePublished - Sep 24 2007
Event21st International Parallel and Distributed Processing Symposium, IPDPS 2007 - Long Beach, CA, United States
Duration: Mar 26 2007Mar 30 2007

Publication series

NameProceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

Other

Other21st International Parallel and Distributed Processing Symposium, IPDPS 2007
CountryUnited States
CityLong Beach, CA
Period3/26/073/30/07

Fingerprint

Checkpoint
File System
Restart
Accumulate
Contention
Tracing
Execution Time
Performance Evaluation
Alignment
Benchmark
Target
Optimization

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Software
  • Mathematics(all)

Cite this

Liao, W. K., Ching, A., Coloma, K., Choudhary, A., & Kandemir, M. (2007). Improving MPI independent write performance using a two-stage write-behind buffering method. In Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM [4228213] (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM). https://doi.org/10.1109/IPDPS.2007.370485
Liao, Wei Keng ; Ching, Avery ; Coloma, Kenin ; Choudhary, Alok ; Kandemir, Mahmut. / Improving MPI independent write performance using a two-stage write-behind buffering method. Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM. 2007. (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM).
@inproceedings{1a6201d0f56f4a788a0ca908cfa7b5b6,
title = "Improving MPI independent write performance using a two-stage write-behind buffering method",
abstract = "Many large-scale production applications often have very long executions times and require periodic data check-points in order to save the state of the computation for program restart and/or tracing application progress. These write-only operations often dominate the overall application runtime, which makes them a good optimization target. Existing approaches for write-behind data buffering at the MPI I/O level have been proposed, but challenges still exist for addressing system-level I/O issues. We propose a two-stage write-behind buffering scheme for handing checkpoint operations. The first-stage of buffering accumulates write data for better network utilization and the second-stage of buffering enables the alignment for the write requests to the file stripe boundaries. Aligned I/O requests avoid file lock contention that can seriously degrade I/O performance. We present our performance evaluation using BTIO benchmarks on both GPFS and Lustre file systems. With the two-stage buffering, the performance of BTIO through MPI independent I/O is significantly improved and even surpasses that of collective I/O.",
author = "Liao, {Wei Keng} and Avery Ching and Kenin Coloma and Alok Choudhary and Mahmut Kandemir",
year = "2007",
month = "9",
day = "24",
doi = "10.1109/IPDPS.2007.370485",
language = "English (US)",
isbn = "1424409101",
series = "Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM",
booktitle = "Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM",

}

Liao, WK, Ching, A, Coloma, K, Choudhary, A & Kandemir, M 2007, Improving MPI independent write performance using a two-stage write-behind buffering method. in Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM., 4228213, Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM, 21st International Parallel and Distributed Processing Symposium, IPDPS 2007, Long Beach, CA, United States, 3/26/07. https://doi.org/10.1109/IPDPS.2007.370485

Improving MPI independent write performance using a two-stage write-behind buffering method. / Liao, Wei Keng; Ching, Avery; Coloma, Kenin; Choudhary, Alok; Kandemir, Mahmut.

Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM. 2007. 4228213 (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM).

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

TY - GEN

T1 - Improving MPI independent write performance using a two-stage write-behind buffering method

AU - Liao, Wei Keng

AU - Ching, Avery

AU - Coloma, Kenin

AU - Choudhary, Alok

AU - Kandemir, Mahmut

PY - 2007/9/24

Y1 - 2007/9/24

N2 - Many large-scale production applications often have very long executions times and require periodic data check-points in order to save the state of the computation for program restart and/or tracing application progress. These write-only operations often dominate the overall application runtime, which makes them a good optimization target. Existing approaches for write-behind data buffering at the MPI I/O level have been proposed, but challenges still exist for addressing system-level I/O issues. We propose a two-stage write-behind buffering scheme for handing checkpoint operations. The first-stage of buffering accumulates write data for better network utilization and the second-stage of buffering enables the alignment for the write requests to the file stripe boundaries. Aligned I/O requests avoid file lock contention that can seriously degrade I/O performance. We present our performance evaluation using BTIO benchmarks on both GPFS and Lustre file systems. With the two-stage buffering, the performance of BTIO through MPI independent I/O is significantly improved and even surpasses that of collective I/O.

AB - Many large-scale production applications often have very long executions times and require periodic data check-points in order to save the state of the computation for program restart and/or tracing application progress. These write-only operations often dominate the overall application runtime, which makes them a good optimization target. Existing approaches for write-behind data buffering at the MPI I/O level have been proposed, but challenges still exist for addressing system-level I/O issues. We propose a two-stage write-behind buffering scheme for handing checkpoint operations. The first-stage of buffering accumulates write data for better network utilization and the second-stage of buffering enables the alignment for the write requests to the file stripe boundaries. Aligned I/O requests avoid file lock contention that can seriously degrade I/O performance. We present our performance evaluation using BTIO benchmarks on both GPFS and Lustre file systems. With the two-stage buffering, the performance of BTIO through MPI independent I/O is significantly improved and even surpasses that of collective I/O.

UR - http://www.scopus.com/inward/record.url?scp=34548738899&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548738899&partnerID=8YFLogxK

U2 - 10.1109/IPDPS.2007.370485

DO - 10.1109/IPDPS.2007.370485

M3 - Conference contribution

AN - SCOPUS:34548738899

SN - 1424409101

SN - 9781424409105

T3 - Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

BT - Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM

ER -

Liao WK, Ching A, Coloma K, Choudhary A, Kandemir M. Improving MPI independent write performance using a two-stage write-behind buffering method. In Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM. 2007. 4228213. (Proceedings - 21st International Parallel and Distributed Processing Symposium, IPDPS 2007; Abstracts and CD-ROM). https://doi.org/10.1109/IPDPS.2007.370485