Live Virtual Machine Migration for In-memory DBs

2022 Fiscal Year Final Research Report

• Project/Area Number: 20K11742
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Review Section: Basic Section 60050:Software-related
• Research Institution: Tokyo University of Agriculture and Technology
• Principal Investigator: Yamada Hiroshi 東京農工大学, 工学(系)研究科(研究院), 准教授 (00571572)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: 仮想マシン移送 / DBMS / インメモリDB

Outline of Final Research Achievements

This project aims at shortening the time for live-migrating virtual machines (VMs) with even large memory DBMSes. Live migration of VMs is a technique that moves active VMs between different physical hosts without losing any running states. Although it is desirable for administrators that the live migration is completed as quickly as possible, the pre-copy-based live migration, widely used in modern hypervisors, does not satisfy this demand on the current trend that VMs on which running applications are performance-critical such as DBMSes have quite large memory. To quickly produce the running state of the migrating VMs on the destination, our approach performs regular memory transfers while simultaneously constructing the DBMS's buffer-pool by fetching the data items from the shared storage. The experimental results show that the migration time of our prototype is shorter under workloads, including sysbench and TPC-C, than the default pre-copy and post-copy schemes, respectively.

Free Research Field

オペレーティングシステム

Academic Significance and Societal Importance of the Research Achievements

当該分野における応募課題の学術的な特色，独創的な点は 1).インメモリDB と VM モニタとを連動させる手法を示している点，2). 提案方式をソフトウェアとして実現する点，3). 提案の有効性を定量的に評価している点などにある．本方式によって，AIサービスのキーであるインメモリDBが稼働する状況においてもVM移送を用いた VM の再配置を可能にし，データセンタの効率的な管理に導く．また，VM モニタとインメモリDBが連動可能であることを示し，新たなデザインスペースの提示することで当該分野の研究領域を活性化していく．