Network with link intermittency is a growing class of emerging networks of interest. Such networks exhibit long data transfer delay caused by discrete communication services. A particular challenge in such networks is the requirement of very large intermediate storage – or “reservoir” – in all the transit elements at their core infrastructure.
In this dissertation, we investigate a forwarding principle called “data flow equilibrium” which aims to substantially reduce transit reservoir consumption compared with conventional Classic IP data forwarding in periodical intermittent networks. We show that this principle can drastically reduce the transit reservoir requirement without degrading the delay. We validate the result in two ways: analytical and simulation.
First, we analytically derive the transit reservoir capacity requirement and transfer delay upper bounds of a single dataflow and critical hop case both, with the equilibrium principle and without it. We devised a data flow equilibrium algorithm, which aims to gear link-hop capacities to preserve transit reservoir.
Second, we validated by simulation the analytical reservoir capacity requirement and transfer delay upper bounds for the general multi-flow case with and without data flow equilibrium. We experiment a data backup drill scenario over a simulated network, whose links are periodically intermittent. We devised two types of Constraint Resource Planning (CRP) routing solvers: Classic and Optimized. Classic solvers are based on the conventional greedy routing and Classis IP forwarding. Optimized solvers are based on scheduling-based intelligent routing and data flow equilibrium forwarding. Simulation results have revealed that data flow equilibrium achieves a near-optimal performance, where the majority of reservoir demands were within their analytical upper bounds. We also shown that data flow equilibrium did not adversely impact data flow transfer delay.
Surprisingly, the presence of data flow equilibrium played the deciding factor in the achievement of perfect completion schedulability. Finally, the most distinguishing aspect of this work is the formal treatment of network intermittency, which, has not been undertaken before.
|Advisor:||Khan, Javed I.|
|School:||Kent State University|
|School Location:||United States -- Ohio|
|Source:||DAI-B 75/08(E), Dissertation Abstracts International|
|Subjects:||Computer Engineering, Electrical engineering, Computer science|
|Keywords:||Data flow, Data forwarding, Equilibrium, Networking, Protocol, Routing|
Copyright in each Dissertation and Thesis is retained by the author. All Rights Reserved
The supplemental file or files you are about to download were provided to ProQuest by the author as part of a
dissertation or thesis. The supplemental files are provided "AS IS" without warranty. ProQuest is not responsible for the
content, format or impact on the supplemental file(s) on our system. in some cases, the file type may be unknown or
may be a .exe file. We recommend caution as you open such files.
Copyright of the original materials contained in the supplemental file is retained by the author and your access to the
supplemental files is subject to the ProQuest Terms and Conditions of use.
Depending on the size of the file(s) you are downloading, the system may take some time to download them. Please be