Real-time systems must meet strict timing requirements to function correctly. These systems are classified as hard or soft, with hard systems having zero tolerance for missed deadlines. Determinism and predictability are crucial for ensuring consistent performance and meeting time constraints.

Timing constraints, including deadlines and timeliness, are fundamental to real-time systems. Jitter and latency must be minimized to maintain reliability. Performance metrics like response time and worst-case execution time are used to analyze and design these systems effectively.

Real-time System Types

Hard and Soft Real-time Systems

• Hard real-time systems have strict timing constraints that must be met to avoid system failure or catastrophic consequences (aircraft control systems, medical devices)
• Soft real-time systems have more flexible timing constraints where occasional deadline misses are tolerable and do not lead to critical failures (multimedia streaming, gaming)
• Hard real-time systems prioritize determinism and predictability over average performance to ensure deadlines are always met
• Soft real-time systems can trade off some determinism for better average-case performance as long as most deadlines are still satisfied

Determinism and Predictability

• Determinism refers to the ability of a system to produce consistent and repeatable results given the same inputs and initial conditions
• Predictability is the ability to accurately estimate or calculate the system's behavior and timing characteristics beforehand
• Real-time systems require a high degree of determinism to ensure that tasks complete within their specified time constraints consistently
• Predictability enables designers to analyze and verify that the system will meet its timing requirements under various conditions (worst-case scenarios, peak loads)

Timing Constraints

Deadlines and Timeliness

• Timeliness is a key requirement for real-time systems, which means tasks must complete within a specified time frame or deadline
• Deadlines can be classified as hard deadlines (missing them leads to system failure) or soft deadlines (occasional misses are tolerable)
• Meeting deadlines consistently is crucial for maintaining the correct functioning and safety of real-time systems (industrial control, robotics)
• Real-time systems must be designed and analyzed to ensure that tasks can complete within their allocated time budgets under all possible conditions

Jitter and Latency

• Jitter refers to the variation in the timing of periodic tasks or events, such as the difference between the expected and actual start or completion times
• Latency is the time delay between the occurrence of an event (input) and the system's response to that event (output)
• Real-time systems aim to minimize jitter and latency to maintain predictable and consistent behavior (sensor data acquisition, actuator control)
• Excessive jitter or latency can lead to missed deadlines, degraded performance, or incorrect system behavior (unstable control loops, delayed responses)

Performance Metrics

Response Time and Worst-Case Execution Time

• Response time is the total time taken by a system to respond to an event, including any waiting time, processing time, and output generation time
• Worst-Case Execution Time (WCET) is the maximum time a task or code segment can take to execute under the worst possible conditions (maximum input size, slowest processor speed)
• Real-time systems are designed and analyzed based on worst-case scenarios to ensure that timing constraints are met even in the most demanding situations
• Techniques like static analysis, measurement-based methods, and probabilistic approaches are used to estimate or calculate WCET for real-time tasks
• Schedulability analysis uses response times and WCETs to determine if a set of tasks can meet their deadlines under a given scheduling policy (Rate Monotonic, Earliest Deadline First)