PROJECT TITLE:

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

ABSTRACT:

This paper presents an initiative knowledge prefetching scheme on the storage servers in distributed file systems for cloud computing. In this prefetching technique, the consumer machines don't seem to be substantially concerned in the method of knowledge prefetching, but the storage servers can directly prefetch the info after analyzing the history of disk I/O access events, and then send the prefetched data to the relevant client machines proactively. To put this technique to work, the knowledge about client nodes is piggybacked onto the $64000 consumer I/O requests, and then forwarded to the relevant storage server. Next, 2 prediction algorithms have been proposed to forecast future block access operations for guiding what knowledge should be fetched on storage servers in advance. Finally, the prefetched data can be pushed to the relevant shopper machine from the storage server. Through a series of evaluation experiments with a assortment of application benchmarks, we tend to have demonstrated that our presented initiative prefetching technique can benefit distributed file systems for cloud environments to attain better I/O performance. In particular, configuration-limited consumer machines within the cloud aren't accountable for predicting I/O access operations, which can positively contribute to preferable system performance on them.