Adaptive provisioning for mobile cloud gaming at edges

Mobile cloud gaming (MCG), which is proposed to deliver high-quality gaming experience to users anywhere and anytime, suffers from tremendous wide-area traffic and long network delays. In order to shorten the delays and provide the gaming services in close proximity to end users, the mobile edge computing (MEC) is envisioned as a promising approach to enable related computations at the edge of the network. Unfortunately, the performance of MEC-enabled MCG highly depends on the provisioning of both gaming services and bandwidth, since the quality of gaming experience (QoE) is easily impaired by both delays and video frame rates. Furthermore, due to the erratic mobility of users, migrating the services accordingly actually decreases the QoE impairment, but incurs extra system cost, leading to the performance-cost tradeoff. In order to address these challenges, we jointly investigate adaptive service placement and bandwidth allocation for MEC-enabled MCG. Considering the system dynamics, we propose to minimize the QoE impairment in a long-term scope, under the constrained cost for migrations. We design an online two-layer iterative algorithm (OnTrial) to solve the problem. Theoretical analysis demonstrates that OnTrial achieves a near-optimal performance and the potential violation of the migration constraint is bounded. Extensive experiments with real-world traces confirm that OnTrial outperforms other algorithms regarding the long-term QoE impairment for games.