云计算中基于缓存优化的镜像模板高效传输方案

Efficient Transmission Scheme of Image Templates Based on Caching Optimization in Cloud Computing

摘要:

在软件即服务的云传输模型中,托管中心根据需求在服务器上部署一个虚拟机镜像模板；镜像模板往往被维护于中央仓库中。由于托管中心物理上离散分布,而且互联网的带宽较低,所以从仓库中传输千兆字节规模的模板文件时存在较大延时；文章利用模板文件间存在的公共性,提出一种ETSIT-CO算法以确定补丁和模板文件在缓存中的分布,通过计算缓存的构成来实现从仓库开始的传输时间最小化,进而降低请求的响应时间；仿真实验结果表明,与标准的模板文件存储缓存技术相比,使用基于补丁的缓存策略实现了很大的性能提升；部署于文章原型测试床上时,基于ETSIT-CO的补丁缓存策略相对只缓存模板的ETSIT-CO选择策略,性能上升90%。

Abstract:

In Software-as-a-Service cloud delivery model, a hosting center deploys a Virtual Machine (VM) image template on a server on demand. Image templates are usually maintained in a central repository. With geographically dispersed hosting centers, time to transfer a large, often GigaByte sized, template file from the repository faces high latency due to low Internet bandwidth. We exploit the presence of this commonality among template files, ETSIT-CO algorithm is proposed to determine the distribution of patched and template files in cache, that computes a cache composition with the objective of minimizing transfers from repository, thereby reducing request service time. Our simulation experiments show significant gains when using patch based caching approach, when compared to standard caching technique of storing template files. Patch-based caching using ETSIT-CO shows over 90% improvement when compared to ETSIT-CO generated selection for template only caching in our prototype testbed implementation.