ML-CLOCK: Efficient Page Cache Algorithm Based on Perceptron-Based Neural Network
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Cho, Minseon | - |
dc.contributor.author | Kang, Donghyun | - |
dc.date.accessioned | 2023-03-13T04:40:05Z | - |
dc.date.available | 2023-03-13T04:40:05Z | - |
dc.date.created | 2023-03-13 | - |
dc.date.issued | 2021-10 | - |
dc.identifier.issn | 2079-9292 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/87051 | - |
dc.description.abstract | Today, research trends clearly confirm the fact that machine learning technologies open up new opportunities in various computing environments, such as Internet of Things, mobile, and enterprise. Unfortunately, the prior efforts rarely focused on designing system-level input/output stacks (e.g., page cache, file system, block input/output, and storage devices). In this paper, we propose a new page replacement algorithm, called ML-CLOCK, that embeds single-layer perceptron neural network algorithms to enable an intelligent eviction policy. In addition, ML-CLOCK employs preference rules that consider the features of the underlying storage media (e.g., asymmetric read and write costs and efficient write patterns). For evaluation, we implemented a prototype of ML-CLOCK based on trace-driven simulation and compared it with the traditional four replacement algorithms and one flash-friendly algorithm. Our experimental results on the trace-driven environments clearly confirm that ML-CLOCK can improve the hit ratio by up to 72% and reduces the elapsed time by up to 2.16x compared with least frequently used replacement algorithms.</p> | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | MDPI | - |
dc.relation.isPartOf | ELECTRONICS | - |
dc.title | ML-CLOCK: Efficient Page Cache Algorithm Based on Perceptron-Based Neural Network | - |
dc.type | Article | - |
dc.type.rims | ART | - |
dc.description.journalClass | 1 | - |
dc.identifier.wosid | 000713775800001 | - |
dc.identifier.doi | 10.3390/electronics10202503 | - |
dc.identifier.bibliographicCitation | ELECTRONICS, v.10, no.20 | - |
dc.description.isOpenAccess | Y | - |
dc.identifier.scopusid | 2-s2.0-85117048803 | - |
dc.citation.title | ELECTRONICS | - |
dc.citation.volume | 10 | - |
dc.citation.number | 20 | - |
dc.contributor.affiliatedAuthor | Kang, Donghyun | - |
dc.type.docType | Article | - |
dc.subject.keywordAuthor | clean-first eviction | - |
dc.subject.keywordAuthor | learning and prediction | - |
dc.subject.keywordAuthor | page replacement algorithm | - |
dc.subject.keywordAuthor | single-layer perceptron neural network | - |
dc.subject.keywordAuthor | sequential write pattern | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalResearchArea | Physics | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Information Systems | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.relation.journalWebOfScienceCategory | Physics, Applied | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
1342, Seongnam-daero, Sujeong-gu, Seongnam-si, Gyeonggi-do, Republic of Korea(13120)031-750-5114
COPYRIGHT 2020 Gachon University All Rights Reserved.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.