Energy Efficient Dynamic Core Allocation for Video Decoding in Embedded Multicore Architectures

This paper describes a dynamic core allocation technique for video decoding on embedded multicore platforms with the objective of reducing energy consumption while guaranteeing a quality of service (QoS). While decoding a frame, the scheme measures "slack" and "overshoot" over the budgeted decode time and amortises across the neighboring frames to achieve overall QoS, compensating for the overshoot with the slack time. It allocates, on a per-frame basis, an appropriate number of cores for decoding to guarantee QoS, while saving energy by using clock gating to switch off unused cores. Using the Sniper simulator to evaluate the implementation of the scheme on a modern embedded processor, we get an energy saving of 6% to 61% while strictly adhering to the required QoS. Thus, we show that substantial energy savings can be achieved in video decoding by employing dynamic core allocation, compared with the default strategy of allocating as many cores as available.