Integrated power management for video streaming to mobile handheld devices

Abstract

Optimizing user experience for streaming video applications on handheld devices is a significant research challenge. In this paper, we propose an integrated power management approach that unifies low level architectural optimizations (CPU, memory, register), OS power-saving mechanisms (Dynamic Voltage Scaling) and adaptive middleware techniques (admission control, optimal transcoding, network traffic regulation). Specifically, we identify interaction parameters between the different levels and optimize them to significantly reduce power consumption. With knowledge of device configurations, dynamic device parameters and changing system conditions, the middleware layer selects an appropriate video quality and fine tunes the architecture for optimized delivery of video. Our performance results indicate that architectural optimizations that are cognizant of user level parameters(e.g. transcoded video quality) can provide energy gains as high as 57.5% for the CPU and memory. Middleware adaptations to changing network noise levels can save as much as 70% of energy consumed by the wireless network interface. Furthermore, we demonstrate how such an integrated framework, that supports tight coupling of inter-level parameters can enhance user experience on a handheld substantially.

A survey of techniques for improving energy efficiency in embedded computing systems

“…Microarchitectural techniques for saving energy in specific components e.g. main memory ( [82][83][84]), cache [6,16,35,[85][86][87][88][89][90][91][92][93][94][95][96][97][98][99][100][101][102][103], scratchpad memory [104], TLB [105] or making other changes to memory hierarchy e.g. adding extra components [106,107].…”

“…Mohapatra et al [35] present a power management technique which integrates microarchitectural level, OS level and middleware level schemes for saving energy in mobile handheld devices. At microarchitectural level, they use static profiling to explore various cache configurations to find the energy-optimal configuration.…”

Markov decision process (MDP) framework for software power optimization using call profiles on mobile phones

“…For example, in [6], the optimal action to turn on/off a disk does not change over time, but in this work optimal decisions depend explicitly on the current time and remaining energy. Quality versus resource utilization trade-offs have also been studied extensively in the area of video streaming [3,9,14,15,[18][19][20]. This is somewhat orthogonal to our work, since most streaming applications would be user-initiated and therefore not delay tolerant.…”

“…For example in [14,15,19,20] researchers present a cross-layer optimization methodology for video decoding by dynamically scaling the voltage and frequency of the underlying processor, such that the task of decoding a given video frame finishes just in time to meet the time requirement. This time constraint can be predicted based on the time required to process a frame derived on the previous history.…”

Chameleon: Application-Level Power Management

“…Their goal is to extend battery lifetime by limiting the average discharge rate and to share this limited resource among competing tasks according to user preferences. An cooperative power management approach was proposed in [20] to unify low level architectural optimizations (CPU, memory, register), OS power-saving mechanisms (Dynamic Voltage and Frequency Scaling) and adaptive middle techniques (admission control, optimal transcoding, network traffic regulation). In this technique, interaction parameters between the different levels are identified and optimized to significantly reduce power consumption.…”