Towards Enhancing QoE for Software Defined Networks Based Cloud Gaming Services

Abstract

The high-profit digital gaming industry has merged with the increasing interest in transforming everything into cloud services, which leads to a novel concept called on-demand gaming. Recently, on-demand gaming, otherwise known as Cloud gaming, has become a promising paradigm for game users and providers. In this thesis, we aim to investigate the optimization of Quality of Experience (QoE) for cloud gaming systems, while considering network challenges, system constraints, and service requirements. We aim to improve the gamer’s QoE from two main network perspectives—cloud perspective, where mechanisms for QoE enhancement must be observed and controlled by the cloud to optimally allocate data center resources to user requested gaming sessions, and home gateways perspective, where the network acquires gamers’ QoE parameters and uses them to maximize the overall QoE of the clients who are sharing the same gateway at home. First, we address the issue of enhancing the client’s QoE by proposing a method to optimally allocate data center resources to user requested gaming sessions. The method considers the current status of the network in terms of communication delay, available computational resources in game servers and processing delay, and the genre of the requested game to select the appropriate server in the cloud for game execution and the network path for game data transfer within the data center. Second, we propose a novel mechanism to optimize the bandwidth distribution in the player’s home network to maximize gaming QoE without starving other concurrent applications. The Proposed method is a home-side Game-Aware Resource Manager (GARM) that intelligently assigns network resources to active applications based on their requirements.