Real-Time Systems Scheduling

Real-time systems scheduling concerns how a computer guarantees that critical tasks—controlling a car's brakes, managing a pacemaker, or coordinating an aircraft's flight surfaces—complete their computations before hard deadlines, where a missed deadline can be as catastrophic as a wrong answer. Researchers study scheduling algorithms that assign processor time and shared resources across tasks, paired with worst-case execution time (WCET) analysis that provides provable upper bounds on how long any piece of code can take to run on specific hardware. As embedded systems increasingly use multicore processors and must simultaneously handle tasks of varying safety importance—a discipline called mixed-criticality scheduling—open questions center on how to maintain timing guarantees without either wasting hardware capacity or allowing low-priority workloads to starve. Predicting execution time on modern processors with caches, pipelines, and speculative execution remains particularly difficult, making hardware-software co-analysis an active area of research.

Works

50,630

Total citations

474,607

Keywords

Scheduling AlgorithmsHard Real-Time SystemsMultiprocessor SchedulingWCET AnalysisEmbedded SystemsTiming Analysis