Reliable Parallel File System Using RAID Technology

Abstract

Providing data avaliability in a cluster environment is very important. Most clusters either use RAID technology or redudant nodes to ensure reliability. High performance clusters usually use parallel file systems to increase throughput. However, when parallel file system is involved in a cluster system, some mechanism must be provided to overcome the side-effect of using striping. PVFS is a popular and open source parallel file system implementation in the Linux environment, but provides no fault tolerence. We propose an idea of using distributed RAID technology to ensure the reliability of using striping. We also introduce a concept called delay write to overcome the write performance penalty incurred in the traditional RAID technology. The evaluation of our RPVFS(Reliable Parallel Virtual File System) shows that the read performance is almost the same when compared with that of original PVFS design. As to the write performance. there only exists a little performance degradation.