PROJECT TITLE:

Performing Initiative Data Prefetching in Distributed File Systems for Cloud Computing - 2015

ABSTRACT:

This paper presents an initiative data prefetching theme on the storage servers in distributed file systems for cloud computing. During this prefetching technique, the client machines don't seem to be substantially involved in the process of knowledge prefetching, however the storage servers can directly prefetch the data when analyzing the history of disk I/O access events, and then send the prefetched data to the relevant client machines proactively. To place this system to work, the knowledge concerning shopper nodes is piggybacked onto the $64000 client I/O requests, and then forwarded to the relevant storage server. Next, two prediction algorithms have been proposed to forecast future block access operations for directing what information ought to be fetched on storage servers earlier. Finally, the prefetched information will be pushed to the relevant consumer machine from the storage server. Through a series of evaluation experiments with a collection of application benchmarks, we tend to have demonstrated that our presented initiative prefetching technique can profit distributed file systems for cloud environments to attain higher I/O performance. In specific, configuration-restricted client machines in the cloud don't seem to be responsible for predicting I/O access operations, which will undoubtedly contribute to preferable system performance on them.