Real-time Systems Overview

Principles

Definition

Real-time systems is a research discipline dedicated to the design, analysis, and implementation of computing systems where the correctness of operations is contingent upon adherence to strict temporal constraints. This field investigates principles and methods to ensure predictable and timely execution, focusing on scheduling, verification, and fault tolerance for applications where delayed responses have significant consequences.

Ontological type

Scheduling Algorithms

Verification Techniques

Core Concepts

Key developments

Key Figures

Formal Timing & Scheduling era

Lui Sha [1] was associated with the University of Pittsburgh [3] and Carnegie Mellon University [4] during the Formal Timing & Scheduling era. His key contributions include the paper Priority inheritance protocols: an approach to real-time synchronization [5] and the paper Aperiodic task scheduling for Hard-Real-Time systems [6], which advanced formal timing guarantees by enabling predictable synchronization and support for non-periodic tasks in hard real-time systems. John P. Lehoczky [2] was affiliated with Carnegie Mellon University [4] during this era. His contributions include the two papers Priority inheritance protocols: an approach to real-time synchronization [5] and Aperiodic task scheduling for Hard-Real-Time systems [6], laying the groundwork for fixed-priority theory and worst-case response-time analysis that enabled safe and schedulable real-time systems in this era.

Energy-Aware Multiprocessor Scheduling era

Lui Sha [1] is associated with Carnegie Mellon University [3] and the University of Illinois Urbana-Champaign [4] during the Energy-Aware Multiprocessor Scheduling era (2000-2009). Sha's key contributions include The rate monotonic scheduling algorithm: exact characterization and average case behavior [6], and On task schedulability in real-time control systems [7], which advanced predictable hard-deadline performance under energy constraints. John P. Lehoczky [2] is affiliated with Carnegie Mellon University [3] and Worcester Polytechnic Institute [5] during this era. Lehoczky's contributions include The rate monotonic scheduling algorithm: exact characterization and average case behavior [6], and On task schedulability in real-time control systems [7], which provided foundational schedulability analyses and energy-aware insights for real-time multiprocessor design.

End-to-End Deterministic Real-Time era

James H. Anderson [1] is a prominent researcher whose activity in this era centers on the University of North Carolina at Chapel Hill [3] and the University of North Carolina at Charlotte [4]. His contributions include Scheduling and locking in multiprocessor real-time operating systems [6], Mixed-Criticality Real-Time Scheduling for Multicore Systems [7], and GPU Scheduling on the NVIDIA TX2: Hidden Details Revealed [8], which collectively strengthened end-to-end determinism by clarifying how scheduling, memory contention, and GPU management shape timing bounds. Sanjoy Baruah [2] is affiliated with Washington University in St. Louis [5] and the University of North Carolina at Chapel Hill [3] during this era. His key results include Response-Time Analysis for Mixed Criticality Systems [9], Mixed-Criticality Real-Time Scheduling for Multicore Systems [7], and An Algorithm for Scheduling Certifiable Mixed-Criticality Sporadic Task Systems [10], which collectively established foundational analysis and scheduling methods for mixed-criticality multicore real-time systems.