Real-Time Systems Scheduling

Real-time systems scheduling concerns how a processor or set of processors can be guaranteed to complete tasks before their deadlines, a property critical in embedded systems ranging from automotive brakes to medical devices where a missed deadline can cause physical harm. Researchers develop and analyze scheduling algorithms—rules that decide which task runs when—and pair them with worst-case execution time (WCET) analysis, which attempts to bound, with mathematical certainty, how long a piece of code can possibly take on a given hardware platform. As systems grow more complex, two challenges dominate current work: multiprocessor scheduling, where tasks must be distributed across many cores without violating timing guarantees, and mixed-criticality systems, where safety-critical and less critical software share the same hardware and must be isolated so that a misbehaving low-priority component cannot jeopardize a high-priority one. Tighter integration of modern processor features—caches, pipelines, speculative execution—makes WCET analysis increasingly difficult to compute precisely, pushing researchers toward new hardware designs and analysis methods that can provide tighter, more trustworthy bounds.

Works

50,297

Total citations

473,471

Keywords

Scheduling AlgorithmsHard Real-Time SystemsMultiprocessor SchedulingWCET AnalysisEmbedded SystemsTiming Analysis