An Optimized Data Duplication Strategy for Federated Clouds using Bloom Filters

Abstract

In Cloud computing deduplication falls under IAAS model as it directly deals with the storage in the datacenters. Deduplication has been implemented at various levels such as file level, block level and chunk level as per the organizational requirements. In the due course of time, many companies have offered for providing better services and subsequently interoperability has braught many Cloud Service Providers (CSP) for scalable provisioning of services under variable workloads, resources and network conditions as hybrid clouds. This concept was very beneficial for customers as well as CSPs and hence it has further evolved and brought many CSPs, either private or public, under one group/alliance and referred it as Federated cloud. Despite of vast storage offered in federated cloud structure, much of it is exhausted with redundant copies of data which needs to be made distinct.  In this paper, we have rendered an optimized deduplication strategy for federated cloud environment in which the domain of the storage in not limited to just one CSP but all the CSPs under one federation.  It has been observed that the work pertaining to this is not available and in this optimized deduplication strategy we have used file level deduplication technique for the implementation of this approach.