%0 Journal Article %1 noauthororeditor %A A.Rasik, %A Anand, S. Raj %D 2014 %E AIRCC, %J The International Journal of Computational Science, Information Technology and Control Engineering (IJCSITCE) %K Acknowledgement,Traffic Computing, Elimination Network Optimization. Predictive Redundancy System,Caching,Cloud coding, identification %N 1 %P 10 %T PREDICTION BASED CLOUD BANDWIDTHAND COSTREDUCTION SYSTEM OF CLOUD COMPUTING %U http://airccse.com/ijcsitce/papers/1114ijcsitce04.pdf %V 1 %X In this paper, we present PACK (Predictive ACKs), a novel destination-to-destination traffic redundancy elimination (TRE) system, Itespecially designed for cloud computing customers. PACK’s main advantage is its capability of offloading the cloud-server TRE effort to end clients, thus reducing the processing costs induced by the TRE algorithm. Not like earlier solutions, PACK is based on a novel TRE technique, which allows the client to use newly received chunks to identify previously received chunk chains, which in turn can be used as reliable predictors to future transmitted chunks.PACK need not require the server to continuously maintain clients’ status. This makes PACK very comfortable for pervasive computation network environments that combine client mobility and server migration to maintain cloud elasticity.Cloudrelated TRE needs to apply a judicious use of cloud resources so that the bandwidth cost reduction combined with the extra cost of TRE computation and data storage would be optimized. We present a new fully functional PACK implementation, transparent to all TCP protocol based applications and all network devices. Finally, we analyze and implement PACK benefits for cloud users, using traffic traces from various sources.

@article{noauthororeditor, abstract = {In this paper, we present PACK (Predictive ACKs), a novel destination-to-destination traffic redundancy elimination (TRE) system, Itespecially designed for cloud computing customers. PACK’s main advantage is its capability of offloading the cloud-server TRE effort to end clients, thus reducing the processing costs induced by the TRE algorithm. Not like earlier solutions, PACK is based on a novel TRE technique, which allows the client to use newly received chunks to identify previously received chunk chains, which in turn can be used as reliable predictors to future transmitted chunks.PACK need not require the server to continuously maintain clients’ status. This makes PACK very comfortable for pervasive computation network environments that combine client mobility and server migration to maintain cloud elasticity.Cloudrelated TRE needs to apply a judicious use of cloud resources so that the bandwidth cost reduction combined with the extra cost of TRE computation and data storage would be optimized. We present a new fully functional PACK implementation, transparent to all TCP protocol based applications and all network devices. Finally, we analyze and implement PACK benefits for cloud users, using traffic traces from various sources. }, added-at = {2017-11-30T06:04:23.000+0100}, author = {A.Rasik and Anand, S. Raj}, biburl = {https://www.bibsonomy.org/bibtex/28acf225a576c599bf800f38adddd5359/suzan}, editor = {AIRCC}, interhash = {e1bf409181461205b9db2810c61f20a7}, intrahash = {8acf225a576c599bf800f38adddd5359}, issn = {ISSN : 2394 - 7527}, journal = {The International Journal of Computational Science, Information Technology and Control Engineering (IJCSITCE) }, keywords = {Acknowledgement,Traffic Computing, Elimination Network Optimization. Predictive Redundancy System,Caching,Cloud coding, identification}, month = {April}, number = 1, pages = 10, timestamp = {2017-11-30T06:04:23.000+0100}, title = {PREDICTION BASED CLOUD BANDWIDTHAND COSTREDUCTION SYSTEM OF CLOUD COMPUTING }, url = {http://airccse.com/ijcsitce/papers/1114ijcsitce04.pdf}, volume = 1, year = { 2014 } }