BROWSE
Detail
Conference Paper
Performance Evaluation of a Distributed File System for VDI Services
Cited 0 time in 
Share 
Share
- Issue Date
- 2014-05
- Citation
- International Conference on Information Technology and Computer Science 2014, pp.1-14
- Language
- English
- Type
- Conference Paper
- Abstract
- Recently VDI technology has been one of hot IT topics owing to many benefits such as security, energy-saving and so on. However it has been slow in its proliferation yet because of several key limitations. Storage is known as the most critical performance bottleneck point and the highest portion of cost at the installation phase. Storage has roughly three issues to be solved. The first issue is the high installation cost which may require high financial burdens to system providers. The second issue is storage isolation problem which SAN as a strong candidate for VDI storage has inherently. The last issue is related with system performance. In this paper, we have implemented a distributed file system, called VDI-FS, to overcome efficiently these limitations. VDI-FS is a scalable, high-performance distributed file system which supports well the characteristics of a VDI system. It makes a single file system with commercial servers with storage devices, instead of expensive SAN storage, and provides system scalability by the function of seamless storage expansion. Moreover in order to improve system performance based on our file system, we introduced a caching function between VM hosts and the file system. We measured performance of the system with IOmark which is widely used to measure storage performance in VDI environment in the absence of standardized tools. It measures average response time with workload specified to VDI system and therefore can be used for VDI system sizing. Our experimental results showed that it is possible for our implementation to satisfy functions and performance required for VDI services with reasonable cost.
- KSP Keywords
- Average response time, Distributed File system, Energy saving, High performance, Performance evaluation, Reasonable Cost, Storage device, System performance, System scalability, installation cost, performance based