English

Abstract

During the last few years there has been a tremendous and efficient improvement in the computational world and one of the area that is leaving its footprints is cloud computing. One of the main trending issue in this computing is traffic redundancy elimination (TRE).Cloud-based TRE needs to apply a judicious use of cloud resources so that the bandwidth cost reduction is optimized together with the additional cost of TRE computation . Here, I present an approach that is suitable for cloud computing customers called Hybrid PACK(predictive acknowledgements) regarding TRE systems. Unlike other techniques PACK’s main advantage is its capability of offloading the cloud-server TRE effort to the end clients ,and hence minimizing the overhead and processing costs induced by the TRE algorithm. In the present scenario the server keeps track of all the end clients. But this is not the case with the PACK as in this mechanism a client can manage his own status and hence the server is offloaded. This makes the PACK suitable for pervasive computing .PACK is based on a novel TRE technique ,which allows the client as well as server to use newly received chunks to identify previously received chunk chains, which in turn can be used as reliable predictors to future transmitted chunks. I here present a hybrid implementation of the PACK in which there is an active involvement of client and server and moreover the chunking signatures are done using SHA 2 (secure hash algorithm) in order to process the any size of chunks.