A Modified Least-Laxity-First scheduling algorithm for real-time tasks
Sung-Heun Oh, Seung‐Min Yang · 2002
The Least-Laxity-First (LLF) scheduling algorithm assigns higher priority to a task with the least laxity, and has been proved to be optimal for uniprocessor systems. The algorithm, however is impractical to implement because laxity tie results in the frequent context switches among the tasks. The Modified Least-Laxity-First (MLLF) scheduling algorithm proposed in this paper solves the problem of the LLF scheduling algorithm by reducing the number of context switches significantly. By reducing the system overhead due to unnecessary context switches, the MLLF scheduling algorithm avoids the degradation of system performance and conserves more system resources for unanticipated aperiodic tasks. We propose the MLLF scheduling algorithm and prove its optimality. We show the performance enhancement of the proposed MLLF scheduling algorithm by using simulation results.