I/O scheduling for limited-size burst-buffers deployed high performance computing

Benbo Zha; Hong Shen

doi:10.1109/PDCAT46702.2019.00021

I/O scheduling for limited-size burst-buffers deployed high performance computing

Benbo Zha, Hong Shen

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

2 Citations (Scopus)

Abstract

Burst-Buffers is a high throughput, small size intermediate storage system integrated between computing nodes and permanent storage system to mitigate the I/O bottleneck problem in modern High Performance Computing (HPC) platforms. This system, however, is unable to effectively handle variable-intensity I/O bursts resulted by unpredictable concurrent accesses to the shared Parallel File System (PFS). In this paper, we introduce a probabilistic I/O scheduling method that takes into account of the burst-buffer load state and instantaneous I/O load distribution of the system based on the probabilistic model of applications to relieve the I/O congestion when I/O load exceeds the PFS bandwidth caused by dynamic application interference. The proposed scheduling method for limited-size Burst-Buffers deployed HPC platforms makes online decision of probabilistic selection of concurrent I/O requests for going through (to PFS), buffering (to Burst-Buffers) or declination in accordance to both the available I/O bandwidth and the current buffer state in order to maximize system efficiency or minimize application dilation. Extensive experiment results on actual characteristic synthetic data show that our method handles the I/O congestion effectively.

Original language	English
Title of host publication	Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019
Editors	Hui Tian, Hong Shen, Wee Lum Tan
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	52-57
Number of pages	6
ISBN (Electronic)	9781728126166
DOIs	https://doi.org/10.1109/PDCAT46702.2019.00021
Publication status	Published - Dec 2019
Externally published	Yes
Event	20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019 - Gold Coast, Australia Duration: 5 Dec 2019 → 7 Dec 2019

Publication series

Name	Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019

Conference

Conference	20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019
Country/Territory	Australia
City	Gold Coast
Period	5/12/19 → 7/12/19

Keywords

Burst-buffers
High performance computing
I/O congestion
I/O scheduling

Access to Document

10.1109/PDCAT46702.2019.00021

Cite this

Zha, B., & Shen, H. (2019). I/O scheduling for limited-size burst-buffers deployed high performance computing. In H. Tian, H. Shen, & W. L. Tan (Eds.), Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019 (pp. 52-57). Article 9028989 (Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/PDCAT46702.2019.00021

Zha, Benbo ; Shen, Hong. / I/O scheduling for limited-size burst-buffers deployed high performance computing. Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019. editor / Hui Tian ; Hong Shen ; Wee Lum Tan. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 52-57 (Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019).

@inproceedings{3565e602a5124043b40d97fdeb497c51,

title = "I/O scheduling for limited-size burst-buffers deployed high performance computing",

abstract = "Burst-Buffers is a high throughput, small size intermediate storage system integrated between computing nodes and permanent storage system to mitigate the I/O bottleneck problem in modern High Performance Computing (HPC) platforms. This system, however, is unable to effectively handle variable-intensity I/O bursts resulted by unpredictable concurrent accesses to the shared Parallel File System (PFS). In this paper, we introduce a probabilistic I/O scheduling method that takes into account of the burst-buffer load state and instantaneous I/O load distribution of the system based on the probabilistic model of applications to relieve the I/O congestion when I/O load exceeds the PFS bandwidth caused by dynamic application interference. The proposed scheduling method for limited-size Burst-Buffers deployed HPC platforms makes online decision of probabilistic selection of concurrent I/O requests for going through (to PFS), buffering (to Burst-Buffers) or declination in accordance to both the available I/O bandwidth and the current buffer state in order to maximize system efficiency or minimize application dilation. Extensive experiment results on actual characteristic synthetic data show that our method handles the I/O congestion effectively.",

keywords = "Burst-buffers, High performance computing, I/O congestion, I/O scheduling",

author = "Benbo Zha and Hong Shen",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019 ; Conference date: 05-12-2019 Through 07-12-2019",

year = "2019",

month = dec,

doi = "10.1109/PDCAT46702.2019.00021",

language = "English",

series = "Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "52--57",

editor = "Hui Tian and Hong Shen and Tan, {Wee Lum}",

booktitle = "Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019",

address = "United States",

}

Zha, B & Shen, H 2019, I/O scheduling for limited-size burst-buffers deployed high performance computing. in H Tian, H Shen & WL Tan (eds), Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019., 9028989, Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019, Institute of Electrical and Electronics Engineers Inc., pp. 52-57, 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019, Gold Coast, Australia, 5/12/19. https://doi.org/10.1109/PDCAT46702.2019.00021

I/O scheduling for limited-size burst-buffers deployed high performance computing. / Zha, Benbo; Shen, Hong.
Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019. ed. / Hui Tian; Hong Shen; Wee Lum Tan. Institute of Electrical and Electronics Engineers Inc., 2019. p. 52-57 9028989 (Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019).

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

TY - GEN

T1 - I/O scheduling for limited-size burst-buffers deployed high performance computing

AU - Zha, Benbo

AU - Shen, Hong

PY - 2019/12

Y1 - 2019/12

N2 - Burst-Buffers is a high throughput, small size intermediate storage system integrated between computing nodes and permanent storage system to mitigate the I/O bottleneck problem in modern High Performance Computing (HPC) platforms. This system, however, is unable to effectively handle variable-intensity I/O bursts resulted by unpredictable concurrent accesses to the shared Parallel File System (PFS). In this paper, we introduce a probabilistic I/O scheduling method that takes into account of the burst-buffer load state and instantaneous I/O load distribution of the system based on the probabilistic model of applications to relieve the I/O congestion when I/O load exceeds the PFS bandwidth caused by dynamic application interference. The proposed scheduling method for limited-size Burst-Buffers deployed HPC platforms makes online decision of probabilistic selection of concurrent I/O requests for going through (to PFS), buffering (to Burst-Buffers) or declination in accordance to both the available I/O bandwidth and the current buffer state in order to maximize system efficiency or minimize application dilation. Extensive experiment results on actual characteristic synthetic data show that our method handles the I/O congestion effectively.

AB - Burst-Buffers is a high throughput, small size intermediate storage system integrated between computing nodes and permanent storage system to mitigate the I/O bottleneck problem in modern High Performance Computing (HPC) platforms. This system, however, is unable to effectively handle variable-intensity I/O bursts resulted by unpredictable concurrent accesses to the shared Parallel File System (PFS). In this paper, we introduce a probabilistic I/O scheduling method that takes into account of the burst-buffer load state and instantaneous I/O load distribution of the system based on the probabilistic model of applications to relieve the I/O congestion when I/O load exceeds the PFS bandwidth caused by dynamic application interference. The proposed scheduling method for limited-size Burst-Buffers deployed HPC platforms makes online decision of probabilistic selection of concurrent I/O requests for going through (to PFS), buffering (to Burst-Buffers) or declination in accordance to both the available I/O bandwidth and the current buffer state in order to maximize system efficiency or minimize application dilation. Extensive experiment results on actual characteristic synthetic data show that our method handles the I/O congestion effectively.

KW - Burst-buffers

KW - High performance computing

KW - I/O congestion

KW - I/O scheduling

UR - http://www.scopus.com/inward/record.url?scp=85083218629&partnerID=8YFLogxK

U2 - 10.1109/PDCAT46702.2019.00021

DO - 10.1109/PDCAT46702.2019.00021

M3 - Conference contribution

AN - SCOPUS:85083218629

T3 - Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019

SP - 52

EP - 57

BT - Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019

A2 - Tian, Hui

A2 - Shen, Hong

A2 - Tan, Wee Lum

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019

Y2 - 5 December 2019 through 7 December 2019

ER -

Zha B, Shen H. I/O scheduling for limited-size burst-buffers deployed high performance computing. In Tian H, Shen H, Tan WL, editors, Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 52-57. 9028989. (Proceedings - 2019 20th International Conference on Parallel and Distributed Computing, Applications and Technologies, PDCAT 2019). doi: 10.1109/PDCAT46702.2019.00021

I/O scheduling for limited-size burst-buffers deployed high performance computing

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this