基于虚拟机页缓存机制的磁盘在线迁移系统  

Live Storage Migration System Based on the Page Cache of the Virtual Machine

在线阅读下载全文

作  者:周志强 李京[1] ZHOU Zhiqiang;LI Jing(School of Computer Science and Technology,University of Science and Technology of China,Hefei 230001,China)

机构地区:[1]中国科学技术大学计算机科学与技术学院,合肥230001

出  处:《小型微型计算机系统》2025年第1期235-241,共7页Journal of Chinese Computer Systems

基  金:中国科学院战略性先导科技专项项目(XDA19020102)资助。

摘  要:在线迁移是云数据中心管理虚拟机的关键技术之一,它在负载均衡、降低能耗以及平台软硬件维护等场景有着重要的作用.预拷贝是目前虚拟机在线迁移的主流方式,有着停机时间短、对虚拟机性能影响小以及鲁棒性高的特点.然而,由于数据的修改会导致多轮迭代传输,该方法在迁移时可能出现较大的网络传输量和较长的迁移时间的问题.考虑到磁盘通常比内存大几个数量级,该问题在迁移磁盘数据时尤为严重.本文通过提取页缓存数据信息实现了一个预拷贝优化方法,并且设计了一个新的磁盘在线迁移系统,使得虚拟机能够参与其自身迁移过程,让获取虚拟机内的信息更加方便.该方法能够高效地进行预拷贝迁移,从而快速而可靠地完成磁盘迁移,减少网络传输数据量和迁移时间.Live migration is one of the key technologies of virtual machine management in cloud data center,which plays an important role in many scenarios such as load balancing,energy management and transparent infrastructure maintenance.Pre-Copy is one of the mainstream live migration methods,which has the characteristics of short downtime,little impact on the performance of virtual machine and high robustness.However,since the modification of data will lead to multiple rounds of iterative transmission,this method may encounter the problem of large network transmission traffic and long migration time.This problem is particularly acute when migrating storage data,given that storage is typically several orders of magnitude larger than memory.This paper implements a pre-copy optimization method by extracting page cache information,and designs a new live storage migration system.Within this migration system,the virtual machine itself can participate in the migration process,which makes it more convenient to obtain the information within the virtual machine.This method can performance the pre-copy migration more efficiently,so as to complete the storage migration quickly and reliably,and reduce the amount of network transmission traffic and migration time.

关 键 词:虚拟机 磁盘在线迁移 预拷贝 页缓存 

分 类 号:TP311[自动化与计算机技术—计算机软件与理论]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象