Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory

Jinfeng Yang; Bingzhe Li; Jianjun Yuan; Zhaoyan Shen; David H Du; David Lilja

doi:10.1109/CASES55004.2022.00010

Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory

Jinfeng Yang, Bingzhe Li, Jianjun Yuan, Zhaoyan Shen, David H Du, David Lilja

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

1 Scopus citations

Abstract

As emerging memory technologies (e.g., non-volatile memory (NVM)) coming out and machine learning algorithms successfully applying to different fields, the potentials of cache replacement policy for NVM-based systems with the integration of machine learning algorithms are worthy of being exploited to improve the performance of computer systems. In this work, we proposed a machine learning based cache replacement algorithm, named ExpCache, to improve the system performance with NVM as the main memory. By considering the non-volatility characteristic of the NVM devices, we split the whole NVM into two caches, including a read cache and a write cache, for retaining different types of requests. The pages in each cache are managed by both LRU and LFU policies for balancing the recency and frequency of workloads. The online Expert machine learning algorithm is responsible for selecting a proper policy to evict a page from one of the caches based on the access patterns of workloads. In experimental results, the proposed ExpCache outperforms previous studies in terms of hit ratio and the number of dirty pages written back to storage.

Original language	English (US)
Title of host publication	Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	7-8
Number of pages	2
ISBN (Electronic)	9781665472968
DOIs	https://doi.org/10.1109/CASES55004.2022.00010
State	Published - 2022
Event	2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022 - Shanghai, China Duration: Oct 7 2022 → Oct 14 2022

Publication series

Name	Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022

Conference

Conference	2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022
Country/Territory	China
City	Shanghai
Period	10/7/22 → 10/14/22

Bibliographical note

Funding Information:
V. ACKNOWLEDGEMENT This work was partially supported by NSF I/UCRC Center Research in Intelligent Storage and the following NSF awards 1439622, 1812537, 2204656, and 2204657. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF.

Publisher Copyright:
© 2022 IEEE.

Keywords

Cache policy
Non-volatile memory
Online learning

Access

10.1109/CASES55004.2022.00010

OpenUrl availability

Full text

Cite this

Yang, J., Li, B., Yuan, J., Shen, Z., Du, D. H., & Lilja, D. (2022). Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory. In Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022 (pp. 7-8). (Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CASES55004.2022.00010

Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory. / Yang, Jinfeng; Li, Bingzhe; Yuan, Jianjun et al.
Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022. Institute of Electrical and Electronics Engineers Inc., 2022. p. 7-8 (Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022).

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

Yang, J, Li, B, Yuan, J, Shen, Z, Du, DH & Lilja, D 2022, Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory. in Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022. Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022, Institute of Electrical and Electronics Engineers Inc., pp. 7-8, 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022, Shanghai, China, 10/7/22. https://doi.org/10.1109/CASES55004.2022.00010

Yang J, Li B, Yuan J, Shen Z, Du DH, Lilja D. Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory. In Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022. Institute of Electrical and Electronics Engineers Inc. 2022. p. 7-8. (Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022). doi: 10.1109/CASES55004.2022.00010

Yang, Jinfeng ; Li, Bingzhe ; Yuan, Jianjun et al. / Work-in-Progress : ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory. Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022. Institute of Electrical and Electronics Engineers Inc., 2022. pp. 7-8 (Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022).

@inproceedings{f86462e468e14a5dad4b38576f3d28eb,

title = "Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory",

abstract = "As emerging memory technologies (e.g., non-volatile memory (NVM)) coming out and machine learning algorithms successfully applying to different fields, the potentials of cache replacement policy for NVM-based systems with the integration of machine learning algorithms are worthy of being exploited to improve the performance of computer systems. In this work, we proposed a machine learning based cache replacement algorithm, named ExpCache, to improve the system performance with NVM as the main memory. By considering the non-volatility characteristic of the NVM devices, we split the whole NVM into two caches, including a read cache and a write cache, for retaining different types of requests. The pages in each cache are managed by both LRU and LFU policies for balancing the recency and frequency of workloads. The online Expert machine learning algorithm is responsible for selecting a proper policy to evict a page from one of the caches based on the access patterns of workloads. In experimental results, the proposed ExpCache outperforms previous studies in terms of hit ratio and the number of dirty pages written back to storage.",

keywords = "Cache policy, Non-volatile memory, Online learning",

author = "Jinfeng Yang and Bingzhe Li and Jianjun Yuan and Zhaoyan Shen and Du, {David H} and David Lilja",

note = "Funding Information: V. ACKNOWLEDGEMENT This work was partially supported by NSF I/UCRC Center Research in Intelligent Storage and the following NSF awards 1439622, 1812537, 2204656, and 2204657. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022 ; Conference date: 07-10-2022 Through 14-10-2022",

year = "2022",

doi = "10.1109/CASES55004.2022.00010",

language = "English (US)",

series = "Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022",

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

pages = "7--8",

booktitle = "Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022",

}

TY - GEN

T1 - Work-in-Progress

T2 - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022

AU - Yang, Jinfeng

AU - Li, Bingzhe

AU - Yuan, Jianjun

AU - Shen, Zhaoyan

AU - Du, David H

AU - Lilja, David

N1 - Funding Information: V. ACKNOWLEDGEMENT This work was partially supported by NSF I/UCRC Center Research in Intelligent Storage and the following NSF awards 1439622, 1812537, 2204656, and 2204657. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the NSF. Publisher Copyright: © 2022 IEEE.

PY - 2022

Y1 - 2022

N2 - As emerging memory technologies (e.g., non-volatile memory (NVM)) coming out and machine learning algorithms successfully applying to different fields, the potentials of cache replacement policy for NVM-based systems with the integration of machine learning algorithms are worthy of being exploited to improve the performance of computer systems. In this work, we proposed a machine learning based cache replacement algorithm, named ExpCache, to improve the system performance with NVM as the main memory. By considering the non-volatility characteristic of the NVM devices, we split the whole NVM into two caches, including a read cache and a write cache, for retaining different types of requests. The pages in each cache are managed by both LRU and LFU policies for balancing the recency and frequency of workloads. The online Expert machine learning algorithm is responsible for selecting a proper policy to evict a page from one of the caches based on the access patterns of workloads. In experimental results, the proposed ExpCache outperforms previous studies in terms of hit ratio and the number of dirty pages written back to storage.

AB - As emerging memory technologies (e.g., non-volatile memory (NVM)) coming out and machine learning algorithms successfully applying to different fields, the potentials of cache replacement policy for NVM-based systems with the integration of machine learning algorithms are worthy of being exploited to improve the performance of computer systems. In this work, we proposed a machine learning based cache replacement algorithm, named ExpCache, to improve the system performance with NVM as the main memory. By considering the non-volatility characteristic of the NVM devices, we split the whole NVM into two caches, including a read cache and a write cache, for retaining different types of requests. The pages in each cache are managed by both LRU and LFU policies for balancing the recency and frequency of workloads. The online Expert machine learning algorithm is responsible for selecting a proper policy to evict a page from one of the caches based on the access patterns of workloads. In experimental results, the proposed ExpCache outperforms previous studies in terms of hit ratio and the number of dirty pages written back to storage.

KW - Cache policy

KW - Non-volatile memory

KW - Online learning

UR - http://www.scopus.com/inward/record.url?scp=85142360823&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85142360823&partnerID=8YFLogxK

U2 - 10.1109/CASES55004.2022.00010

DO - 10.1109/CASES55004.2022.00010

M3 - Conference contribution

AN - SCOPUS:85142360823

T3 - Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022

SP - 7

EP - 8

BT - Proceedings - 2022 International Conference on Compilers, Architecture, and Synthesis for Embedded Systems, CASES 2022

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 7 October 2022 through 14 October 2022

ER -

Work-in-Progress: ExpCache: Online-Learning based Cache Replacement Policy for Non-Volatile Memory

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access

OpenUrl availability

Other files and links

Fingerprint

Cite this