*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Optimizing of Network Resource Allocation for Software-Defined Data Center Networks
Committee:
Dr. Douglas Blough, ECE, Chair , Advisor
Dr. Henry Owen, ECE
Dr. Yusun Chang, ECE
Dr. Linda Wills, ECE
Dr. Ellen Zegura, CoC
Abstract:
As cloud computing and data center network flourishes, the network that was once designed to support traditional networking scenario must now satisfy new requirements to suit for the cloud environment and increasing demands. The Software-Defined Networking (SDN) paradigm, with the control plane separated from the data plane, is widely regarded as the next generation networking technique. The objective of this thesis is to optimize network resources allocation in the software-defined data center networks (DCN). The SDN resources considered here are the SDN switch to controller link bandwidth and the switch flow table size. First, a queueing model is developed to provision the SDN switches with an appropriate number of switch-to-controller connections. Second, a controller-level admission control mechanism is proposed to determine if a new flow should be admitted to the network when the flow table is congested. Third, we study the fair and high-satisfaction resources allocation problem with the routing path optimized in software-defined DCN. The delay guarantees for delay-sensitive flows are also provided. Finally, some practical issues are considered for the resources allocation algorithms. The provided theoretical analysis and simulation results in this dissertation improve the efficiency of resource allocation in software-defined DCN.
