首页 | 本学科首页   官方微博 | 高级检索  
     

虚拟机陷出的检测及分析
引用本文:汪小林,张彬彬,靳辛欣,王振林,罗英伟,李晓明. 虚拟机陷出的检测及分析[J]. 计算机科学与探索, 2011, 5(6): 493-500
作者姓名:汪小林  张彬彬  靳辛欣  王振林  罗英伟  李晓明
作者单位:1. 北京大学信息科学技术学院,北京,100871
2. 密西根理工大学计算机系,美国密西根州49931-1295
基金项目:60873052,the National Grand Basic Research 973 Program of China under Grant No.2007CB310900(国家重点基础研究发展规划,the National High-Tech Research and Development of China under Grant No.2008AA01Z112(国家高技术研究发展计划,the MOE-Intel Information Technology Foundation under Grant No.MOE-INTEL-10-06
摘    要:虚拟机执行敏感指令时会陷出到虚拟机管理器(virtual machine monitor,VMM)处理,虚拟机频繁陷出是影响虚拟化性能的重要因素,因此全面了解导致陷出的敏感指令对虚拟化性能优化有重要意义。提出了一个创新性的方法"桶竞争法"(competition in bucket method,CBM),通过把敏感指令的地址映射到不同的桶中,采用竞争方式在各个桶内寻找陷出次数最多的几个地址,能高效地跟踪所有的虚拟机陷出。

关 键 词:虚拟化  虚拟机管理器(VMM)  热指令  虚拟机陷出  桶竞争法(CBM)
修稿时间: 

Detection and Analysis of Virtual Machine Exits
WANG Xiaolin,ZHANG Binbin,JIN Xinxin,WANG Zhenlin,LUO Yingwei,LI Xiaoming. Detection and Analysis of Virtual Machine Exits[J]. Journal of Frontier of Computer Science and Technology, 2011, 5(6): 493-500
Authors:WANG Xiaolin  ZHANG Binbin  JIN Xinxin  WANG Zhenlin  LUO Yingwei  LI Xiaoming
Affiliation:1. School of Electronics Engineering and Computer Science, Peking University, Beijing 100871, China 2. Department of Computer Science, Michigan Technological University, Michigan 49931-1295, USA
Abstract:It’s essential to learn about sensitive instructions that cause virtual machine (VM) to exit to the virtual machine monitor (VMM) to find new ways to optimize the performance of VM. This paper proposes a novel method, competition in bucket method (CBM), to track all VM exits efficiently, by mapping sensitive instructions to buckets according to instruction addresses, and find out hot instructions in each bucket.
Keywords:virtualization  virtual machine monitor (VMM)  hot instructions  virtual machine exits  competition in bucket method (CBM)
本文献已被 CNKI 万方数据 等数据库收录!
点击此处可从《计算机科学与探索》浏览原始摘要信息
点击此处可从《计算机科学与探索》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号