(SEM VI) THEORY EXAMINATION 2023-24 REAL TIME SYSTEMS

REAL TIME SYSTEMS – KOE061

Section-wise Important Questions & Ready Answers

SECTION A

(Attempt all – 2 marks each)

(a) Hard vs Soft Real-Time Systems

In hard real-time systems, missing a deadline is catastrophic and can cause system failure, such as in avionics or medical devices. In soft real-time systems, missing a deadline degrades performance but does not lead to total failure, as seen in multimedia streaming or online gaming.

(b) Definition of a Real-Time System

A real-time system is a computing system in which correctness depends not only on logical results but also on the time at which results are produced. Meeting timing constraints is essential for system reliability.

(c) Dynamic vs Static Systems in Scheduling

In static systems, task parameters such as priorities and execution times are known beforehand and remain fixed. In dynamic systems, task attributes can change at runtime, requiring adaptive scheduling decisions.

(d) Scheduling of Aperiodic and Sporadic Jobs

Aperiodic and sporadic jobs are scheduled using priority-driven approaches such as background scheduling, polling servers, or deferrable servers to ensure that periodic task deadlines are not violated.

(e) Resource Access Control (RAC)

Resource Access Control ensures safe and predictable access to shared resources in real-time systems. It prevents issues such as priority inversion and ensures deadline compliance.

(f) Techniques for Concurrent Data Access

Techniques include mutex locks, semaphores, priority inheritance protocols, priority ceiling protocols, and non-preemptive critical sections to manage concurrent access safely.

(g) Soft vs Hard Real-Time Communication Systems

Hard real-time communication requires guaranteed delivery within strict deadlines, while soft real-time communication tolerates delays and occasional losses without system failure.

(h) Function of MAC Protocols

Medium Access Control (MAC) protocols regulate access to shared communication media, ensuring fairness, collision avoidance, and predictable data transmission in broadcast networks.

(i) POSIX Issues in Real-Time Systems

POSIX real-time issues include scheduling latency, priority handling, timer resolution, interrupt response, and support for real-time threads and synchronization primitives.

(j) Commercial Real-Time Databases

Commercial real-time databases support deadline-aware transactions, temporal data, concurrency control, and predictable response times, often used in telecom and industrial control systems.

SECTION B

(Attempt any three – 10 marks each)

2(a) Importance of Release Times, Deadlines, and Timing Constraints

Release times define when a task becomes ready, deadlines specify completion limits, and timing constraints ensure system predictability. These factors directly affect scheduling feasibility and determine whether a real-time system can guarantee deadline satisfaction.

2(b) Weighted Round Robin Scheduling

Weighted Round Robin allocates processor time based on predefined weights, ensuring fairness while honoring task importance. It is simpler than priority-based scheduling but less effective for strict deadline guarantees.

2(c) Effects of Resource Contention

Resource contention causes blocking, increased response times, and missed deadlines. It impacts system predictability and requires protocols like priority inheritance or priority ceiling to mitigate negative effects.

2(d) Real-Time Communication over the Internet

Challenges include variable delays, packet loss, and bandwidth limitations. Resource reservation protocols like RSVP allocate bandwidth and maintain Quality of Service (QoS) for real-time applications such as video conferencing.

2(e) UNIX as a Real-Time Operating System

UNIX requires enhancements like preemptive kernels, real-time scheduling classes, high-resolution timers, and predictable interrupt handling to support real-time applications. While flexible, it has limitations in strict hard real-time environments.

SECTION C

3(a) Periodic Task Model

In the periodic task model, tasks execute repeatedly at fixed intervals. Each task is characterized by execution time, period, and deadline. This model simplifies schedulability analysis and resource allocation.

3(b) Impact of Missing Deadlines

In hard real-time systems, missing deadlines may cause catastrophic failure. In soft real-time systems, it results in reduced performance, degraded output quality, or temporary instability.

4(a) Optimality of EDF and LST Algorithms

EDF dynamically schedules tasks based on nearest deadlines and is optimal for uniprocessor systems. LST schedules tasks with minimum slack time, offering flexibility but higher computational overhead.

4(b) Rate Monotonic Algorithm (RMA)

RMA assigns fixed priorities based on task periods. Shorter periods get higher priority. It is simple and predictable but less efficient than EDF in processor utilization.