Open Collections

Optimizing cloud gaming service delivery

University of British Columbia

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 cloud gaming. In this thesis, we aim to investigate the optimization of quality of experience (QoE) for cloud gaming system, while considering different challenges, system constraints and service requirements. First, we investigate video compression technologies based on existing cloud gaming system, in which the cloud hosts the game engine and streams rendered gaming videos to players through the Internet. We propose to cooperatively encode cloud gaming videos of different players in the same game session, in order to leverage inter-gamer redundancy. This is based on an observation that game scenes of close-by gamers have non-trivial overlapping areas, and thus adding inter-gamer predictive video frames may improve the coding efficiency. Selected games are analyzed and the trace-driven simulations demonstrate the efficiency of proposed system. Second, we introduce a novel decomposed cloud gaming paradigm, which supports flexible migrations of gaming components between the cloud server and the players' terminals. We present the blueprint of the proposed system and discussed the cognitive resource optimization for the proposed decomposed cloud gaming system under distinct targets. This includes the minimization of cloud, network, and terminal resources and response delay, subject to QoE assurance, which is formulated as a graph partitioning problem that is solved by exhaustive searches. Extensive simulation results show the feasibility of cognitive resource management in a cloud gaming system to efficiently adapt itself to variations in the service environments, while satisfying different QoE requirements for gaming sessions. Finally, we explore the practical approach for the decomposed cloud gaming paradigm. We design the system framework and seek the engineering solutions for practical issues. Following these discussions, we implement the very first experimental testbed called ubiquitous cloud gaming platform. Three game prototypes are built on our testbed, which can demonstrate the feasibility and efficiency of our proposal. Experiments have been conducted to show that intelligent partitioning leads to better system performance, such as lower response latency and higher frame rate.

Text

2016-04-25

Attribution-NonCommercial-NoDerivatives 4.0 International

http://hdl.handle.net/2429/57861

Electrical and Computer Engineering

University of British Columbia

UBCV

http://creativecommons.org/licenses/by-nc-nd/4.0/