Computational Load and Congestion Control in Cloud Environments
Author(s): F. Kottmann |
Conference/Journal: Master Thesis, FS15 (10470) |
Abstract: Today, the allocation of computational power for distributed computing over the internet is not fexible regarding changing bandwidth constraints. Due to the raising importance of cloud computing and infrastructure as a service, the first markets are being implemented to exchange computational power. However, the optimal allocation of resources remains an open problem in that context. In this thesis, a model is introduced for the problem of optimal allocation of the computing power and of the corresponding data ows, according to bandwidth and computing capacity constraints, for a given network graph and routing structure. It is shown that the two stage interaction and the anticipation of the Internet congestion avoidance mechanisms is a bilevel optimization problem, which is generally non convex and hard to solve, but its convexification obtains the same unique optimal solution. Supervisors: Saverio Bolognani, Florian Dörfler | Year: 2015 |
Type of Publication: (12)Diploma/Master Thesis | |
Supervisor: F. Dörfler | |
% Autogenerated BibTeX entry @PhdThesis { Xxx:2015:IFA_5317 } |