Power Management for Mobile GPUs

Abstract

The graphics processing units (GPUs) in mobile devices have come a long way in terms of their capabilities to accelerate both graphics- and general-purpose applications. However, they also consume a significant amount of power when executing such applications. This necessitates sophisticated power management techniques for the GPUs in order to save power and energy in mobile devices that typically have limited power budget and battery capacity.In this chapter, we first discuss the advantages and limitations of the state-of-the-art power management techniques for both gaming- and general-purpose applications running on GPUs. In particular, existing approaches suffer from lack of coordination in power management between the central processing unit (CPU) and the GPU coinciding on the same chip in the mobile application processors. We present our recently proposed power management solutions to address these shortcomings through synergistic CPU-GPU execution and power management. Finally, we identify and elaborate on the open problems that provide further opportunities for reduction in power consumption of mobile GPUs.

Publication
Elsevier Advances in GPU Research and Practice
Anuj Pathania
Anuj Pathania
Assistant Professor

Anuj Pathania is an Assistant Professor in the Parallel Computing Systems (PCS) group at the University of Amsterdam (UvA). His research focuses on the design of sustainable systems deployed in power-, thermal-, energy- and reliability-constrained environments.