Frame-Based Dynamic Voltage and Frequency Scaling for an MPEG Player

This paper describes a dynamic voltage and frequency scaling (DVFS) technique for MPEG decoding to reduce the energy consumption while maintaining a quality of service (QoS) constraint. The computational workload for an incoming frame is predicted by using a frame-based history so that the processor voltage and frequency can be scaled to provide the exact amount of computing power needed to decode the frame. More precisely, the required decoding time for each frame is separated into two parts: a frame-dependent (FD) part and a frame-independent (FI) part. The FD part varies greatly according to the type of the incoming frame whereas the FI part remains constant regardless of the frame type. Separation of the FI part from the overall decoding sequence provides two key benefits depending on the hardware platform: better compensation of the error due to workload prediction and higher level of energy saving when given a QoS degradation level. The proposed DVFS scheme has been implemented on two platforms, a low performance StrongArm-1110-based evaluation board from Intel and a high performance XScale-based testbed designed at USC. In the StrongArm-1110-based system, the FI part is used to compensate for the prediction error that may occur during the FD part, whereas in the XScale-based system, the FI part is used to reduce energy consumption by employing the lowest CPU frequency during the corresponding time intervals. Detailed current measurements in these two platforms demonstrate larger than 87% and 80% CPU energy saving while maintaining a user-provided frame, respectively. Index Terms — Dynamic voltage and frequency scaling, MPEG decoding, workload decomposition.