基于堆叠文件系统的云计算存储优化  

Optimization of Cloud Computing Storage Based on Stackable File System

作　　者：郭婉[1] 张晓[1] 丰文雄 

机构地区：[1]西北工业大学计算机学院,西安710129

出　　处：《计算机工程》2014年第12期19-25,共7页Computer Engineering

基　　金：国家"863"计划基金资助重大项目(2013AA01A215);国家自然科学基金资助项目(61033007);西北工业大学基础研究基金资助项目(JC20120209)

摘　　要：云环境中大量虚拟镜像带来的冗余会造成资源利用率降低,采用重复数据删除技术能缓解对存储空间以及网络带宽的需求,但同时也存在处理性能下降的问题。为此,提出一种基于堆叠文件系统的云计算存储优化方案。该方案利用可堆叠文件系统将用户对虚拟实例的少量修改信息存放于可读写盘中,将虚拟镜像存放于只读盘中,每次只保存用户对虚拟实例的少量修改信息,而无需保存整个实例。实验结果表明,该方案可在不影响性能的条件下降低存储容量,对文件系统性能的影响平均约为0.75%,在用户对虚拟实例只需少量修改的情况下,对存储容量的需求相比传统重复数据删除方案减少约35%。The vast virtual images in the cloud environment brings a lot of redundancy which makes massive reduce of resource utilization. Although the use of data deduplication technology can ease the demand for storage space and network bandwidth,it faces processing performance degradation. This paper presents a solution based on stackable file system to optimize storage in cloud environment. Using stackable file system to store a small amount of modifies on virtual instances in read/write disk,the virtual images are stored in a read-only disk. It saves a small amount of modification of virtual instances and does not have to keep the whole instance to save. Experimental results show that this scheme can reduce the storage capacity and almost do not affect the performance,and the effect on the performance of the file system is only about 0. 75% on average. In the case of user virtual instances with only minor modifications,the program can reduce demand for storage capacity by about 35% compared with traditional deduplication solutions.

关 键 词：堆叠文件系统 虚拟图像 云计算 联合文件系统 重复数据删除 

分 类 号：TP333[自动化与计算机技术—计算机系统结构]