Scalaraid: Optimizing linux software raid system for next-generation storage

Shushu Yi; Yanning Yang; Yunxiao Tang; Zixuan Zhou; Junzhe Li; Chen Yue; Myoungsoo Jung; Jie Zhang

doi:10.1145/3538643.3539740

Scalaraid: Optimizing linux software raid system for next-generation storage

Shushu Yi, Yanning Yang, Yunxiao Tang, Zixuan Zhou, Junzhe Li, Chen Yue, Myoungsoo Jung, Jie Zhang

School of Integrated Technology

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

Abstract

RAID has been widely adopted to enhance the performance, capacity, and reliability of the existing storage systems. However, we observe that the Linux software RAID (mdraid) suffers from its poor implementation of the lock mechanism. To address this, we propose ScalaRAID, which refines the role domain of locks and designs a new data structure to prevent different threads from preempting the RAID resources. By doing so, ScalaRAID can maximize the thread-level parallelism and reduce the time consumption of I/O request handling. Our evaluation results reveal that ScalaRAID can improve throughput by 89.4% while decreasing 99.99th percentile latency by 85.4% compared to mdraid.

Original language	English
Title of host publication	HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems
Publisher	Association for Computing Machinery, Inc
Pages	119-125
Number of pages	7
ISBN (Electronic)	9781450393997
DOIs	https://doi.org/10.1145/3538643.3539740
Publication status	Published - 2022 Jun 27
Event	14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022 - Virtual, Online, United States Duration: 2022 Jun 27 → 2022 Jun 28

Publication series

Name	HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems

Conference

Conference	14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022
Country/Territory	United States
City	Virtual, Online
Period	22/6/27 → 22/6/28

Bibliographical note

Funding Information:
The authors thank Prof. Guangyan Zhang of Tsinghua University for shepherding their paper. This research is mainly supported by Peking University start-up package (7100603645). Dr. Jung is in part supported by NRF 2021R1AC4001773 and IITP 2021-0-00524 & 2022-0-00117, KAIST IDEC & Start-up (G01190015), Samsung HiPHER, and Samsung Research Grant (G01200447). Jie Zhang is the corresponding author.

Publisher Copyright:
© 2022 ACM.

All Science Journal Classification (ASJC) codes

Computer Networks and Communications
Hardware and Architecture
Information Systems
Software

Access to Document

10.1145/3538643.3539740

Cite this

Yi, S., Yang, Y., Tang, Y., Zhou, Z., Li, J., Yue, C., Jung, M., & Zhang, J. (2022). Scalaraid: Optimizing linux software raid system for next-generation storage. In HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems (pp. 119-125). (HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems). Association for Computing Machinery, Inc. https://doi.org/10.1145/3538643.3539740

Yi, Shushu ; Yang, Yanning ; Tang, Yunxiao et al. / Scalaraid : Optimizing linux software raid system for next-generation storage. HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems. Association for Computing Machinery, Inc, 2022. pp. 119-125 (HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems).

@inproceedings{cb611594aa2d4532ac73319f98dfe618,

title = "Scalaraid: Optimizing linux software raid system for next-generation storage",

abstract = "RAID has been widely adopted to enhance the performance, capacity, and reliability of the existing storage systems. However, we observe that the Linux software RAID (mdraid) suffers from its poor implementation of the lock mechanism. To address this, we propose ScalaRAID, which refines the role domain of locks and designs a new data structure to prevent different threads from preempting the RAID resources. By doing so, ScalaRAID can maximize the thread-level parallelism and reduce the time consumption of I/O request handling. Our evaluation results reveal that ScalaRAID can improve throughput by 89.4% while decreasing 99.99th percentile latency by 85.4% compared to mdraid.",

author = "Shushu Yi and Yanning Yang and Yunxiao Tang and Zixuan Zhou and Junzhe Li and Chen Yue and Myoungsoo Jung and Jie Zhang",

note = "Funding Information: The authors thank Prof. Guangyan Zhang of Tsinghua University for shepherding their paper. This research is mainly supported by Peking University start-up package (7100603645). Dr. Jung is in part supported by NRF 2021R1AC4001773 and IITP 2021-0-00524 & 2022-0-00117, KAIST IDEC & Start-up (G01190015), Samsung HiPHER, and Samsung Research Grant (G01200447). Jie Zhang is the corresponding author. Publisher Copyright: {\textcopyright} 2022 ACM.; 14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022 ; Conference date: 27-06-2022 Through 28-06-2022",

year = "2022",

month = jun,

day = "27",

doi = "10.1145/3538643.3539740",

language = "English",

series = "HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems",

publisher = "Association for Computing Machinery, Inc",

pages = "119--125",

booktitle = "HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems",

}

Yi, S, Yang, Y, Tang, Y, Zhou, Z, Li, J, Yue, C, Jung, M & Zhang, J 2022, Scalaraid: Optimizing linux software raid system for next-generation storage. in HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems. HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems, Association for Computing Machinery, Inc, pp. 119-125, 14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022, Virtual, Online, United States, 22/6/27. https://doi.org/10.1145/3538643.3539740

Scalaraid : Optimizing linux software raid system for next-generation storage. / Yi, Shushu; Yang, Yanning; Tang, Yunxiao et al.

HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems. Association for Computing Machinery, Inc, 2022. p. 119-125 (HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems).

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

TY - GEN

T1 - Scalaraid

T2 - 14th ACM Workshop on Hot Topics in Storage and File Systems, HotStorage 2022

AU - Yi, Shushu

AU - Yang, Yanning

AU - Tang, Yunxiao

AU - Zhou, Zixuan

AU - Li, Junzhe

AU - Yue, Chen

AU - Jung, Myoungsoo

AU - Zhang, Jie

N1 - Funding Information: The authors thank Prof. Guangyan Zhang of Tsinghua University for shepherding their paper. This research is mainly supported by Peking University start-up package (7100603645). Dr. Jung is in part supported by NRF 2021R1AC4001773 and IITP 2021-0-00524 & 2022-0-00117, KAIST IDEC & Start-up (G01190015), Samsung HiPHER, and Samsung Research Grant (G01200447). Jie Zhang is the corresponding author. Publisher Copyright: © 2022 ACM.

PY - 2022/6/27

Y1 - 2022/6/27

N2 - RAID has been widely adopted to enhance the performance, capacity, and reliability of the existing storage systems. However, we observe that the Linux software RAID (mdraid) suffers from its poor implementation of the lock mechanism. To address this, we propose ScalaRAID, which refines the role domain of locks and designs a new data structure to prevent different threads from preempting the RAID resources. By doing so, ScalaRAID can maximize the thread-level parallelism and reduce the time consumption of I/O request handling. Our evaluation results reveal that ScalaRAID can improve throughput by 89.4% while decreasing 99.99th percentile latency by 85.4% compared to mdraid.

AB - RAID has been widely adopted to enhance the performance, capacity, and reliability of the existing storage systems. However, we observe that the Linux software RAID (mdraid) suffers from its poor implementation of the lock mechanism. To address this, we propose ScalaRAID, which refines the role domain of locks and designs a new data structure to prevent different threads from preempting the RAID resources. By doing so, ScalaRAID can maximize the thread-level parallelism and reduce the time consumption of I/O request handling. Our evaluation results reveal that ScalaRAID can improve throughput by 89.4% while decreasing 99.99th percentile latency by 85.4% compared to mdraid.

UR - http://www.scopus.com/inward/record.url?scp=85134148452&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85134148452&partnerID=8YFLogxK

U2 - 10.1145/3538643.3539740

DO - 10.1145/3538643.3539740

M3 - Conference contribution

AN - SCOPUS:85134148452

T3 - HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems

SP - 119

EP - 125

BT - HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems

PB - Association for Computing Machinery, Inc

Y2 - 27 June 2022 through 28 June 2022

ER -

Yi S, Yang Y, Tang Y, Zhou Z, Li J, Yue C et al. Scalaraid: Optimizing linux software raid system for next-generation storage. In HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems. Association for Computing Machinery, Inc. 2022. p. 119-125. (HotStorage 2022 - Proceedings of the 2022 14th ACM Workshop on Hot Topics in Storage and File Systems). doi: 10.1145/3538643.3539740

Scalaraid: Optimizing linux software raid system for next-generation storage

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this