Dissertation/Thesis Abstract

Prediction based cloud bandwidth and cost reduction system: PACK
by Koundinyan, Shubha, M.S., California State University, Long Beach, 2016, 32; 10102605
Abstract (Summary)

Predictive Acknowledgments (PACK) is the latest cloud-based end-to-end Traffic Redundancy Elimination system (TRE). Cloud-based systems have to optimize the usage of resources such as bandwidth. The bandwidth usage is reduced by decreasing the cost of TRE computation and storage. PACK is a receiver based TRE system, which uses the newly received data chunks to identify the previously received data chunks. PACK protocol diminishes the cost induced by the TRE algorithm. The server status does not have to be continuously maintained when using the PACK protocol. This feature makes PACK suitable for client mobility and server migration, which creates universal calculation environments responsible for maintaining cloud elasticity. This report presents the design and implementation of PACK protocol. Simulation results show that this design reduces the overall operational cloud cost and minimizes the application latency between the sender and the receiver.

Indexing (document details)
Advisor: Chassiakos, Anastasios
Commitee: Aghnatios, Wajdi, Ary, James, Chassiakos, Anastasios
School: California State University, Long Beach
Department: Electrical Engineering
School Location: United States -- California
Source: MAI 55/04M(E), Masters Abstracts International
Source Type: DISSERTATION
Subjects: Electrical engineering
Keywords: Bandwidth, Cloud, Cost, Pack, Prediction, Reduction
Publication Number: 10102605
ISBN: 9781339662879
Copyright © 2019 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy
ProQuest