Thermal-throttling server: A thermal-aware real-time task scheduling framework for three-dimensional multicore chips

• A thermal-aware real-time scheduling framework for 3D multicore chips is proposed.
• A thermal size assignment is used to manage the heat generated by task executions.
• The thermal-throttling server handles run-time temperature management.
• The thermal-throttling real-time dispatchers are presented.
• The admission control ensures that all tasks meet the timing and thermal constraints.

Three-dimensional (3D) multicore chips have been recently developed to deal with the power consumption and interconnection delay problems of embedded systems; however, thermal management has proven to be challenging due to the heat effect of vertically stacked cores, and the subsequent trade-off that occurs between performance requirements and overheating. In this paper we propose a novel thermal-aware real-time scheduling framework for 3D multicore chips to achieve an effective trade-off between system temperature and task schedulability for dynamic workloads. A thermal-throttling server is first proposed to adjust the heat generated by task executions, and a thermal-throttling dispatcher is then presented to enable thermal-awareness in well-known real-time dispatchers. An admission control is subsequently derived to ensure that all task executions satisfy the thermal and timing constraints. Lastly, a series of extensive simulations are carried out, with encouraging results in terms of schedulability and the prevention of overheating.