Colocation Games and Their Application to Distributed Resource Management

We introduce Colocation Games as the basis of a general framework for modeling, analyzing, and facilitating the interactions between the various stakeholders in distributed/cloud computing environments, where resources are offered in an open marketplace to independent, rational parties interested in setting up their own applications. Virtualization technologies enable the partitioning of such resources so as to allow each player to dynamically acquire appropriate fractions of the resources. When all the components are under the control of a single administrative domain, this leads to an standard optimization problem, but when infrastructure providers make available their resources in a marketplace, and from there customers acquire the resources, the global optimization framework is no longer appropriate. Rather, in this paper we use a game-theoretic framework in which the assignment of players to resources is the outcome of a strategic ”Colocation Game”. Although we show that determining the existence of an equilibrium for colocation games in general is NP-hard, we present a number of simplified, practically-motivated variants of the colocation game for which we establish convergence to a Nash Equilibrium, and price of anarchy bounds. In addition to these analytical results, we present an experimental evaluation of implementations of some of these variants. Experimental evaluation corroborates our analytical results and also illustrates how colocation games offer a feasible distributed resource management alternative for self-organizing systems, in which the adoption of a global optimization approach would be neither practical nor justifiable. † Supported in part by the Universidad Pontificia Bolivariana and COLCIENCIAS–Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnologı́a “Francisco José de Caldas”. ‡ Supported in part by NSF awards CCF-0820138, CSR-0720604, EFRI-0735974, CNS-0524477, and CNS-0520166. ♭ Supported in part by NSF award CCR-0635102.