File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Energy-efficient dynamic virtual machine management in data centers

TitleEnergy-efficient dynamic virtual machine management in data centers
Authors
KeywordsCloud computing
resource management
energy efficiency
Markov decision process
Issue Date2019
PublisherInstitute of Electrical and Electronics Engineers. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=90
Citation
IEEE/ACM Transactions on Networking, 2019, v. 27 n. 1, p. 344-360 How to Cite?
AbstractEfficient virtual machine (VM) management can dramatically reduce energy consumption in data centers. Existing VM management algorithms fall into two categories based on whether the VMs' resource demands are assumed to be static or dynamic. The former category fails to maximize the resource utilization as they cannot adapt to the dynamic nature of VMs' resource demands. Most approaches in the latter category are heuristic and lack theoretical performance guarantees. In this paper, we formulate the dynamic VM management as a large-scale Markov decision process (MDP) problem and derive an optimal solution. Our analysis of real-world data traces supports our choice of the modeling approach. However, solving the large-scale MDP problem suffers from the curse of dimensionality. Therefore, we further exploit the special structure of the problem and propose an approximate MDP-based dynamic VM management method, called MadVM. We prove the convergence of MadVM and analyze the bound of its approximation error. Moreover, we show that MadVM can be implemented in a distributed system with at most two times of the optimal migration cost. Extensive simulations based on two real-world workload traces show that MadVM achieves significant performance gains over two existing baseline approaches in power consumption, resource shortage, and the number of VM migrations. Specifically, the more intensely the resource demands fluctuate, the more MadVM outperforms.
Persistent Identifierhttp://hdl.handle.net/10722/293708
ISSN
2023 Impact Factor: 3.0
2023 SCImago Journal Rankings: 2.034
ISI Accession Number ID

 

DC FieldValueLanguage
dc.contributor.authorHAN, Z-
dc.contributor.authorTan, H-
dc.contributor.authorWang, R-
dc.contributor.authorChen, G-
dc.contributor.authorLI, Y-
dc.contributor.authorLau, FCM-
dc.date.accessioned2020-11-23T08:20:39Z-
dc.date.available2020-11-23T08:20:39Z-
dc.date.issued2019-
dc.identifier.citationIEEE/ACM Transactions on Networking, 2019, v. 27 n. 1, p. 344-360-
dc.identifier.issn1063-6692-
dc.identifier.urihttp://hdl.handle.net/10722/293708-
dc.description.abstractEfficient virtual machine (VM) management can dramatically reduce energy consumption in data centers. Existing VM management algorithms fall into two categories based on whether the VMs' resource demands are assumed to be static or dynamic. The former category fails to maximize the resource utilization as they cannot adapt to the dynamic nature of VMs' resource demands. Most approaches in the latter category are heuristic and lack theoretical performance guarantees. In this paper, we formulate the dynamic VM management as a large-scale Markov decision process (MDP) problem and derive an optimal solution. Our analysis of real-world data traces supports our choice of the modeling approach. However, solving the large-scale MDP problem suffers from the curse of dimensionality. Therefore, we further exploit the special structure of the problem and propose an approximate MDP-based dynamic VM management method, called MadVM. We prove the convergence of MadVM and analyze the bound of its approximation error. Moreover, we show that MadVM can be implemented in a distributed system with at most two times of the optimal migration cost. Extensive simulations based on two real-world workload traces show that MadVM achieves significant performance gains over two existing baseline approaches in power consumption, resource shortage, and the number of VM migrations. Specifically, the more intensely the resource demands fluctuate, the more MadVM outperforms.-
dc.languageeng-
dc.publisherInstitute of Electrical and Electronics Engineers. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=90-
dc.relation.ispartofIEEE/ACM Transactions on Networking-
dc.rightsIEEE/ACM Transactions on Networking. Copyright © Institute of Electrical and Electronics Engineers.-
dc.rights©20xx IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.-
dc.subjectCloud computing-
dc.subjectresource management-
dc.subjectenergy efficiency-
dc.subjectMarkov decision process-
dc.titleEnergy-efficient dynamic virtual machine management in data centers-
dc.typeArticle-
dc.identifier.emailLau, FCM: fcmlau@cs.hku.hk-
dc.identifier.authorityLau, FCM=rp00221-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1109/TNET.2019.2891787-
dc.identifier.pmid30623258-
dc.identifier.scopuseid_2-s2.0-85061708369-
dc.identifier.hkuros319197-
dc.identifier.volume27-
dc.identifier.issue1-
dc.identifier.spage344-
dc.identifier.epage360-
dc.identifier.isiWOS:000458851600025-
dc.publisher.placeUnited States-
dc.identifier.issnl1063-6692-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats