CSR: PDOS: RIVER: Resource Management Infrastructure for Consolidated Hosting in Virtualized Data Centers

最近，服务器虚拟化研究的复苏引起了大型数据中心提供商的极大兴趣，他们希望利用该技术为托管主机设计改进的解决方案。虚拟化有望提高资源整合的程度，同时降低管理、维修和电力的运营成本。因此，从多个方面设计高效的虚拟化数据中心是一项值得尝试的工作。然而，从传统的托管模型到虚拟化模型的转换并不是一件简单的事情。实现虚拟化所提供的与整合相关的好处需要重新考虑：(i) VMM中的调度器，（ii）资源使用监控和记帐机制，以及（iii）系统范围内的动态资源供应机制。该提案确定了虚拟化数据中心在所有这些方面出现的关键挑战，并开发了一个称为River的综合解决方案来解决这些问题。River将做出以下具体贡献：(i)改进虚拟机监视器（VMM）内的资源管理算法，它使用水平（在VMM中的调度程序之间）和垂直（在VMM和托管的操作系统调度程序之间）合作，（ii）高效和稳定的动态资源配置，以及（iii）系统原语，它将为配置算法提供新的机会，通过减少虚拟化开销来提高成本削减。一个基于Xen的River数据中心原型将被开发出来，并将其代码提供给社区。最后，我们将重新设计与宾夕法尼亚州立大学系统课程相关的实验室，以采用虚拟托管的课堂项目。

The recent resurgence of research in server virtualization has created a lot of interest among providers of large-scaledata centers in employing this technology to design improved solutions for managed hosting. Virtualization holds the promise of increaseddegrees of resource consolidation with accompanying reduction inoperational costs of administration, repair, and electricity. Designing efficient virtualized data centers is therefore a desirable and worthy endeavor in multiple ways. The transition from traditional hosting models to a virtualized model is, however, not a trivial one. Realizingthe consolidation-related benefits that virtualization has tooffer requires a re-consideration of: (i) schedulers within the VMM,(ii) mechanisms for resource usage monitoring and accounting, and (iii) system-wide dynamic resource provisioning mechanisms. This proposal identifies the key challenges that arise onall these fronts in a virtualized data center, and develops a comprehensive solution, called River, to address them. River will make the following specific contributions: (i) improvedresource management algorithms within the virtual machine monitor (VMM) that use both horizontal (between the schedulers in the VMM) and vertical (between VMM and hosted OS schedulers) co-operation, (ii) efficient and stable dynamic resource provisioning, and (iii) systems primitives that would provide new opportunitiesto the provisioning algorithms for improved cost-cutting via reduction of virtualization overheads. A prototype River data center basedon Xen will be developed and resulting code made available to the community. Finally, we willre-design the labs associated with the systems courses at Penn State to employ virtualized hosting of class projects.