Datenbestand vom 09. März 2024

Warenkorb Datenschutzhinweis Dissertationsdruck Dissertationsverlag Institutsreihen     Preisrechner

aktualisiert am 09. März 2024

ISBN 9783843911016

84,00 € inkl. MwSt, zzgl. Versand


978-3-8439-1101-6, Reihe Informatik

Ulrich Lampe
Monetary Efficiency in Infrastructure Clouds – Solution Strategies for Workload Distribution and Auction-based Capacity Allocation

162 Seiten, Dissertation Technische Universität Darmstadt (2013), Hardcover, B5

Zusammenfassung / Abstract

Since the early days of computing, a vision has been to provide Information Technology services in the form of a utility, just like water, electricity, or telephony. With the advancement of the cloud computing paradigm since the mid-2000s, this vision has been put into practice. Cloud computing builds on and combines multiple existing technologies and paradigms, such as virtualization and Service-Oriented Architectures, to deliver various forms of Information Technology services over the Internet. In this thesis, our focus is on the most elementary class of Information Technology: computing infrastructure. In this context, we examine two important research problems and propose solution strategies, based on the conjoint objective of monetary efficiency.

As the first major contribution, we introduce the so-called Cloud-oriented Workload Distribution Problem (CWDP). This problem concerns the distribution of a workload, which comprises multiple computational jobs, across leased infrastructure. We assume the position of a cloud user, who aims at cost-minimal deployment under consideration of resource constraints. On the basis of a mathematical optimization model, we propose the exact solution approach CWDP-EXA.KOM. Given its high time complexity, we further propose the heuristic optimization approach CWDP-HEU.KOM, which is complemented by the improvement procedure CWDP-IMP.KOM. The practical applicability and performance of these optimization approaches is demonstrated using a quantitative evaluation, based on realistic data from the cloud computing market.

As the second key contribution, we examine the Equilibrium Price Auction Allocation Problem (EPAAP). This problem refers to the allocation of Virtual Machine instances based on an equilibrium price auction scheme. For that matter, we focus on the role of a cloud provider, who pursues the aim of profit maximization. We formalize the problem as an optimization model, which permits to deduce the exact optimization approach EPAAP-EXA.KOM. We further propose a heuristic optimization approach, named EPAAP-HEU.KOM, and improvement procedure EPAAP-IMP.KOM. All three approaches are thoroughly analyzed through a quantitative evaluation.