Loading...
Benchmarking and exploiting resource utilization in cloud gaming: security and privacy challenges
Bui, Sang Liu, Guannan
Bui, Sang
Liu, Guannan
Citations
Altmetric:
Advisor
Editor
Date
Date Issued
2025-04
Date Submitted
Keywords
Collections
Research Projects
Organizational Units
Journal Issue
Embargo Expires
Abstract
Cloud gaming platforms rely on large scale data centers to stream games, minimizing the need for high end local hardware. To reduce costs, these platforms often use overprovisioning, allocating shared server resources under the assumption that not all users will require peak performance simultaneously. However, this strategy introduces vulnerabilities in power consumption and thermal management, where sudden demand spikes may cause overheating, performance degradation, or system failures. This study will benchmark the thermal and power demands of various video games to evaluate these vulnerabilities. Using real time CPU and GPU performance metrics including voltage, temperature, and workload, along with thermal imaging, we will assess how typical gaming sessions stress cloud infrastructure under standard conditions. We will also investigate whether specific graphical features such as ray tracing and shader enhancements can intentionally increase hardware usage. We will also test this by developing our own modifications to select games and adjusting in game settings, allowing us to evaluate the potential for power and thermal attacks intended to overwhelm server systems. Our findings will identify exploitable resource patterns and expose weaknesses in current data center mitigation strategies. This research will contribute to a deeper understanding of performance based attack vectors in cloud gaming and provide recommendations for improving infrastructure resilience.
Associated Publications
Rights
Copyright of the original work is retained by the author.