Scheduling instructions for enhanced performance and energy efficiency on a clustered architecture with applications in big-data sensor networks

Devices in wireless sensor network are powered by battery, energy consumption of those devices becomes a major concern. So we need to enhance the performance and energy efficiency of those wireless sensor devices. This paper focuses on the research of enhancing the performance and energy efficiency of clustered VLIW DSP architecture served as the core processing unit on the wireless sensor devices. A novel scheduling technology is presented in this work, which exploits the trade-off between balancing the distributions of instructions among clusters and reducing of the amount of inter-cluster data communications to guide the cluster assign and cycle schedule procedure. Results show that large improvement can be achieved in performance and energy efficiency.