Presentation 2018-05-17
A Proposal of Virtual Machine Relocation Method Using Hierarchical Reinforcement Learning
Shohei Enomoto, Tatsuro Kimura, Toyonori Fujiura,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In recent years, the cloud has attracted attention as a platform for analysis of big data and IoT service, and it is expected that the needs of the cloud continue to increase in the future. For this reason, it is one of the major tasks for the cloud provider to increase the number of virtual machines on the physical machine, that is, to improve the aggregation rate. Currently, cloud operators are improving aggregation rate by overcommitting, but under overcommitting environment there is a risk of lowering service level due to resource competition. Since the workload of the virtual machine is constantly fluctuating and the virtual machines are frequently added and deleted, the load between the physical machines tends to be unbalanced, and in a high-load situation, the risk of a decline in the service level is high. In this paper, we propose a virtual machine relocation method using hierarchical reinforcement learning which balances the load among physical machines, avoids contention of resources, and maintains the service level.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Cloud / Virtual Machine / Virtual Machine Relocation Problem / Reinforcement Learning / Load Balancing
Paper # ICM2018-2
Date of Issue 2018-05-10 (ICM)

Conference Information
Committee ICM / IPSJ-CSEC / IPSJ-IOT
Conference Date 2018/5/17(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Kiyohito Yoshihara(KDDI Research)
Vice Chair Yoichi Yamashita(NTT Neomeit) / Takumi Miyoshi(Shibaura Inst. Tech.)
Secretary Yoichi Yamashita(KDDI Research) / Takumi Miyoshi(NTT)
Assistant Yuncheng Zhu(Hitachi)

Paper Information
Registration To Technical Committee on Information and Communication Management / Special Interest Group on Computer Security / Special Interest Group on Internet and Operation Technology
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Proposal of Virtual Machine Relocation Method Using Hierarchical Reinforcement Learning
Sub Title (in English)
Keyword(1) Cloud
Keyword(2) Virtual Machine
Keyword(3) Virtual Machine Relocation Problem
Keyword(4) Reinforcement Learning
Keyword(5) Load Balancing
1st Author's Name Shohei Enomoto
1st Author's Affiliation NIPPON TELEGRAPH AND TELEPHONE CORPORATION(NTT)
2nd Author's Name Tatsuro Kimura
2nd Author's Affiliation NIPPON TELEGRAPH AND TELEPHONE CORPORATION(NTT)
3rd Author's Name Toyonori Fujiura
3rd Author's Affiliation NIPPON TELEGRAPH AND TELEPHONE CORPORATION(NTT)
Date 2018-05-17
Paper # ICM2018-2
Volume (vol) vol.118
Number (no) ICM-39
Page pp.pp.27-32(ICM),
#Pages 6
Date of Issue 2018-05-10 (ICM)