Power Efficient Scheduling for Network on Chip Applications on Multicore Processor

摘要

Task scheduling on multicore systems for optimized power and energy consumption is essential as the trend of utilizing multiple processors is ever increasing. In the design and implementation of today's multicore/many core chips, power and energy have become increasingly important concerns. In this project, we present priority-based CPU scheduling algorithms that use dynamic performance data to reduce the power consumption by over 20% with significant increase in performance. This algorithm operates at different fixed frequency and uses Linux CPU sets and cores. Many technique like dynamic frequency scaling, lower the core's frequency during the execution of the task and hence lowering performance. This algorithm match processes to core better suited to execute those processes in an effort to lower the average completion time of all processes in an entire task, thus improving performance. This algorithms use dynamic process priorities as scheduling criteria for reducing power and energy consumption hence indirectly reduces cores temperatures and aids in eliminating the hot spots to ensure overall thermal safety. In this project, we propose an optimal ILP solution to task scheduling for different applications on a multicore system with power and energy constraints. Effectiveness of our techniques is shown in result on different benchmark.