الفهرس | Only 14 pages are availabe for public view |
Abstract Several real-time embedded system applications demand predictable timing behavior and satisfy other system constraints, such as energy consumption. With the advent of multicore processors in the embedded market, reducing energy consumption is becoming increasingly important for multicore processors as well. The situation is critical in portable devices where the energy budget is limited and hence battery lifetime defines the usefulness of the system. Therefore energy consumption has become a major concern in the design of real-time embedded systems. This thesis addresses the issue of overall energy optimization in real-time embedded systems at the operating system level using efficient real-time task scheduling algorithms; on the Dynamic Voltage Frequency Scaling (DVFS)-capable multicore systems. Initially, a novel workload partitioning algorithm is proposed, namely Blocking-Aware Based Partitioning (BABP). It statically allocates real-time tasks with shared resources to a homogenous multicore processor such that the blocking time of these dependent tasks is significantly reduced. The BABP algorithm effectively exploits the available parallelism, balances the workload in these multicore systems, and assigns tasks that can run in parallel to different cores as much as possible, taking into account blocking-time minimization. The BABP algorithm reduces an average of 10.5%, 39.0%, and 48.7% of energy consumption compared with SBP, WFD, and BFD respectively, for short period task sets, and it reduces an average of 11.13%, 25.23% and 51.44% energy consumption compared with SBP, WFD, and BFD respectively, for long period task sets. |