Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O

Hongchao Du; Qiao Li; Riwei Pan; Tei-Wei Kuo; Chun Jason Xue

doi:10.1109/HPCA56546.2023.10071009

Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O

Hongchao Du, Qiao Li^*, Riwei Pan, Tei-Wei Kuo, Chun Jason Xue

^*Corresponding author for this work

Department of Computer Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

The complex software stack has become the performance bottleneck of the system with high-speed Non-Volatile Memory (NVM). Memory-mapped I/O (MMIO) could avoid the long-stack overhead by bypassing the kernel, but the performance is limited by existing crash-resilient mechanisms. We propose a Multi-Granularity Shadow Paging (MGSP) strategy, which smartly utilizes the redo and undo logs as shadow logs to provide a light-weight crash-resilient mechanism for MMIO. In addition, a multi-granularity strategy is designed to provide high-performance updating and locking for reducing runtime overhead, where strong consistency is preserved with a lockfree metadata log. Experimental results show that the proposed MGSP achieves 1.1 ∼ 4.21× performance improvement with write and 2.56 ∼ 3.76× improvement with multi-threads write compared with the underlying file system. For SQLite, MGSP can improve the database performance by 29.4% for Mobibench and 36.5% for TPCC, on average. ©2023 IEEE.

Original language	English
Title of host publication	2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)
Publisher	IEEE
Pages	108-121
ISBN (Electronic)	978-1-6654-7652-2
ISBN (Print)	978-1-6654-7653-9
DOIs	https://doi.org/10.1109/HPCA56546.2023.10071009
Publication status	Published - 2023
Event	29th IEEE International Symposium on High-Performance Computer Architecture (HPCA 2023) - Montreal, Canada Duration: 25 Feb 2023 → 1 Mar 2023 https://hpca-conf.org/2023/ https://ieeexplore.ieee.org/xpl/conhome/1000335/all-proceedings

Publication series

Name	IEEE Symposium on High-Performance Computer Architecture
ISSN (Print)	1530-0897
ISSN (Electronic)	2378-203X

Conference

Conference	29th IEEE International Symposium on High-Performance Computer Architecture (HPCA 2023)
Country/Territory	Canada
City	Montreal
Period	25/02/23 → 1/03/23
Internet address	https://hpca-conf.org/2023/ https://ieeexplore.ieee.org/xpl/conhome/1000335/all-proceedings

Bibliographical note

Research Unit(s) information for this publication is provided by the author(s) concerned.

Funding

We would like to thank anonymous reviewers for their valuable comments. This work was supported in part by the Research Grants Council of the Hong Kong Special Administrative Region, China under Grant CityU 11209122, and by the National Natural Science Foundation of China under Grant No. 62202396. The corresponding author is Qiao Li ([email protected]).

Access Output

10.1109/HPCA56546.2023.10071009

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Du, H., Li, Q., Pan, R., Kuo, T.-W., & Xue, C. J. (2023). Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O. In 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA) (pp. 108-121). (IEEE Symposium on High-Performance Computer Architecture). IEEE. https://doi.org/10.1109/HPCA56546.2023.10071009

@inproceedings{cf68de78d7664c678ee771e61bb3ee59,

title = "Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O",

abstract = "The complex software stack has become the performance bottleneck of the system with high-speed Non-Volatile Memory (NVM). Memory-mapped I/O (MMIO) could avoid the long-stack overhead by bypassing the kernel, but the performance is limited by existing crash-resilient mechanisms. We propose a Multi-Granularity Shadow Paging (MGSP) strategy, which smartly utilizes the redo and undo logs as shadow logs to provide a light-weight crash-resilient mechanism for MMIO. In addition, a multi-granularity strategy is designed to provide high-performance updating and locking for reducing runtime overhead, where strong consistency is preserved with a lockfree metadata log. Experimental results show that the proposed MGSP achieves 1.1 ∼ 4.21× performance improvement with write and 2.56 ∼ 3.76× improvement with multi-threads write compared with the underlying file system. For SQLite, MGSP can improve the database performance by 29.4% for Mobibench and 36.5% for TPCC, on average. {\textcopyright}2023 IEEE.",

author = "Hongchao Du and Qiao Li and Riwei Pan and Tei-Wei Kuo and Xue, {Chun Jason}",

note = "Research Unit(s) information for this publication is provided by the author(s) concerned.; 29th IEEE International Symposium on High-Performance Computer Architecture (HPCA 2023) ; Conference date: 25-02-2023 Through 01-03-2023",

year = "2023",

doi = "10.1109/HPCA56546.2023.10071009",

language = "English",

isbn = "978-1-6654-7653-9",

series = "IEEE Symposium on High-Performance Computer Architecture",

publisher = "IEEE",

pages = "108--121",

booktitle = "2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)",

address = "United States",

url = "https://hpca-conf.org/2023/, https://ieeexplore.ieee.org/xpl/conhome/1000335/all-proceedings",

}

Du, H , Li, Q , Pan, R, Kuo, T-W & Xue, CJ 2023, Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O. in 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA). IEEE Symposium on High-Performance Computer Architecture, IEEE, pp. 108-121, 29th IEEE International Symposium on High-Performance Computer Architecture (HPCA 2023), Montreal, Quebec, Canada, 25/02/23. https://doi.org/10.1109/HPCA56546.2023.10071009

Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O. / Du, Hongchao ; Li, Qiao ; Pan, Riwei et al.
2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA). IEEE, 2023. p. 108-121 (IEEE Symposium on High-Performance Computer Architecture).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O

AU - Du, Hongchao

AU - Li, Qiao

AU - Pan, Riwei

AU - Kuo, Tei-Wei

AU - Xue, Chun Jason

N1 - Research Unit(s) information for this publication is provided by the author(s) concerned.

PY - 2023

Y1 - 2023

N2 - The complex software stack has become the performance bottleneck of the system with high-speed Non-Volatile Memory (NVM). Memory-mapped I/O (MMIO) could avoid the long-stack overhead by bypassing the kernel, but the performance is limited by existing crash-resilient mechanisms. We propose a Multi-Granularity Shadow Paging (MGSP) strategy, which smartly utilizes the redo and undo logs as shadow logs to provide a light-weight crash-resilient mechanism for MMIO. In addition, a multi-granularity strategy is designed to provide high-performance updating and locking for reducing runtime overhead, where strong consistency is preserved with a lockfree metadata log. Experimental results show that the proposed MGSP achieves 1.1 ∼ 4.21× performance improvement with write and 2.56 ∼ 3.76× improvement with multi-threads write compared with the underlying file system. For SQLite, MGSP can improve the database performance by 29.4% for Mobibench and 36.5% for TPCC, on average. ©2023 IEEE.

AB - The complex software stack has become the performance bottleneck of the system with high-speed Non-Volatile Memory (NVM). Memory-mapped I/O (MMIO) could avoid the long-stack overhead by bypassing the kernel, but the performance is limited by existing crash-resilient mechanisms. We propose a Multi-Granularity Shadow Paging (MGSP) strategy, which smartly utilizes the redo and undo logs as shadow logs to provide a light-weight crash-resilient mechanism for MMIO. In addition, a multi-granularity strategy is designed to provide high-performance updating and locking for reducing runtime overhead, where strong consistency is preserved with a lockfree metadata log. Experimental results show that the proposed MGSP achieves 1.1 ∼ 4.21× performance improvement with write and 2.56 ∼ 3.76× improvement with multi-threads write compared with the underlying file system. For SQLite, MGSP can improve the database performance by 29.4% for Mobibench and 36.5% for TPCC, on average. ©2023 IEEE.

UR - http://www.scopus.com/inward/record.url?scp=85151703873&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85151703873&origin=recordpage

U2 - 10.1109/HPCA56546.2023.10071009

DO - 10.1109/HPCA56546.2023.10071009

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 978-1-6654-7653-9

T3 - IEEE Symposium on High-Performance Computer Architecture

SP - 108

EP - 121

BT - 2023 IEEE International Symposium on High-Performance Computer Architecture (HPCA)

PB - IEEE

T2 - 29th IEEE International Symposium on High-Performance Computer Architecture (HPCA 2023)

Y2 - 25 February 2023 through 1 March 2023

ER -

Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O

Abstract

Publication series

Conference

Bibliographical note

Funding

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Towards Unified-storage-memory-enabled Mobile Devices

Cite this

Multi-Granularity Shadow Paging with NVM Write Optimization for Crash-Consistent Memory-Mapped I/O

Abstract

Publication series

Conference

Bibliographical note

Funding

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Towards Unified-storage-memory-enabled Mobile Devices

Cite this