基于重复数据删除的虚拟桌面存储优化技术

虚拟桌面基础架构依靠数据中心海量的云基础设施,为用户按需提供虚拟桌面部署所需的软硬件资源,但同时面临存储资源利用率低和虚拟机启动慢的困境.针对虚拟桌面存储中具有大量数据冗余的特性,采用重复数据删除技术缩减虚拟桌面基础架构的存储空间需求;并利用服务器本地磁盘缓存以及共享存储池内的固态硬盘来优化虚拟机的启动性能.通过原型实现,发现相比于基于内容分块的策略,静态分块策略更适合虚拟桌面存储进行重复数据删除,最优的分块大小为4KB,并能够缩减85%的存储空间容量;通过服务器本地磁盘缓存和基于闪存的固态硬盘进行I/O优化,虚拟机的启动速度能够获得35%的提升.

Deduplication Based Storage Optimization Technique for Virtual Desktop

Depending on massive cloud infrastructure in data centers, virtual desktop infrastructure (VDI) can provide software and hardware resources for user on demand. However, there are two challenges facing VDI: low storage capacity utilization and low desktop performance. In this paper, we present a deduplication based storage optimization technique for virtual desktop environment. It can significantly reduce storage capacity requirement based on large amounts of data redundancy in virtual desktop storage, and boost desktop performance by server local disk cache and solid state disk (SSD) in hybrid storage pool. The experimental evaluations, based on our prototype implementation, show that static chunking based deduplication can get higher deduplication ratio and efficiency than content defined chunking based deduplication, and 4 KB is its optimal chunk size with 85% storage capacity saving; Our I/O optimiza-tion based on local disk cache and SSD can speedup the virtual machine startup performance by 35%.