THEORY EXAMINATION (SEM–VIII) 2016-17 REAL TIME SYSTEM

Real-Time System – Section Wise Explanation

Section A – Basic Concepts of Real-Time Systems

Section A contains short questions that test the basic understanding of real-time computing systems and scheduling concepts. Students must attempt all questions in this section. 

A real-time system is a computer system that must respond to events within a specific time limit. In such systems, correctness depends not only on logical results but also on the time at which results are produced.

Real-time systems are widely used in applications such as air traffic control systems, medical monitoring systems, industrial automation, automotive systems, and telecommunications.

One concept related to real-time systems is the embedded system. An embedded system is a specialized computer system designed to perform a specific task within a larger system. Many embedded systems operate as real-time systems.

Another important concept is the Target Operating System (TargetOS). This is the operating system running on the target hardware platform where the real-time application executes.

Real-time communication networks also support hard real-time and soft real-time communication. Hard real-time systems must strictly meet deadlines, while soft real-time systems allow occasional deadline misses.

Network concepts such as traffic shaping and policing are used to regulate network traffic and maintain quality of service.

Another important concept is Quality of Service (QoS) routing, which ensures that network resources meet the required performance levels for real-time communication.

Real-time systems often use event-driven schedulers, which schedule tasks based on events such as task arrival or completion.

Understanding these basic concepts helps engineers design reliable real-time systems.

Questions for Section A

What is a real-time system?

What issues occur in real-time system scenarios?

What is an embedded system? How is it different from a real-time system?

What is TargetOS?

What is the difference between open systems and closed systems?

What is the difference between hard real-time and soft real-time communication?

What is the difference between traffic shaping and traffic policing?

What is QoS routing?

Are all hard real-time systems safety-critical?

Why are scheduling decisions made only at task arrival and completion in a non-preemptive scheduler?

Section B – Real-Time Scheduling and System Design

Section B focuses on scheduling algorithms, system constraints, and real-time operating system design issues. Students must attempt any five questions from this section. 

Real-time systems must satisfy two types of constraints:

Performance constraints – requirements related to timing and system performance.

Behavioral constraints – requirements related to correct system behavior.

Scheduling is a critical aspect of real-time systems. One commonly used scheduling algorithm is Earliest Deadline First (EDF). EDF is a dynamic priority scheduling algorithm where tasks with the nearest deadlines are given higher priority.

Another scheduling technique mentioned in the exam paper is the table-driven scheduler. In this approach, scheduling decisions are precomputed and stored in a scheduling table before system execution.

Multiprocessor systems and distributed systems present different scheduling challenges. Algorithms that work well in multiprocessor systems may not work effectively in distributed systems because distributed systems have communication delays and synchronization issues.

Traditional operating systems such as Unix are not ideal for real-time applications because they do not guarantee strict timing constraints.

Dynamic priority scheduling techniques allow task priorities to change during execution. This helps improve scheduling efficiency in some real-time applications.

Concurrency control protocols are also important in real-time databases because multiple tasks may need access to shared resources simultaneously.

The Two-Phase Locking (2PL) protocol is commonly used in database systems, but it is not suitable for real-time databases because it can cause unpredictable delays and missed deadlines.

Understanding these concepts helps engineers design efficient and reliable real-time systems.

Questions for Section B

What is the difference between performance constraints and behavioral constraints?

Is EDF a dynamic priority scheduling algorithm? Explain.

Compute the schedule table length for tasks using a table-driven scheduler.

Why are multiprocessor scheduling algorithms not suitable for distributed systems?

What are the drawbacks of using Unix kernel for real-time applications?

How does dynamic priority scheduling affect real-time systems?

Explain concurrency protocols used in real-time systems.

Why is the traditional Two-Phase Locking protocol unsuitable for real-time databases?

Section C – Advanced Real-Time System Concepts

Section C contains advanced questions related to real-time task classification, clock synchronization, and input-output operations in real-time systems. Students must attempt any two questions from this section. 

Real-time tasks can be classified into three categories:

Periodic tasks – tasks that occur at regular time intervals.

Aperiodic tasks – tasks that occur irregularly without a fixed time interval.

Sporadic tasks – tasks that occur irregularly but have a minimum time interval between occurrences.

Clock synchronization is another important concept in distributed real-time systems. Multiple processors must maintain synchronized clocks to ensure correct execution of time-critical tasks.

There are two main approaches for clock synchronization:

Centralized clock synchronization – one master clock controls all other clocks.

Distributed clock synchronization – multiple clocks cooperate to maintain synchronization.

Each method has advantages and disadvantages related to reliability, accuracy, and fault tolerance.

Another important concept in real-time systems is input/output (I/O) operations. I/O operations can be classified as:

Synchronous I/O – the process waits until the I/O operation completes.

Asynchronous I/O – the process continues execution while the I/O operation occurs in the background.

Asynchronous I/O is often better suited for real-time systems because it allows tasks to execute without waiting for slow I/O operations.

Understanding these advanced concepts helps engineers design efficient and reliable real-time computing systems.

Questions for Section C

What are periodic, aperiodic, and sporadic real-time tasks?

Why is clock synchronization required in distributed real-time systems?

Compare centralized and distributed clock synchronization.

What is the difference between synchronous and asynchronous I/O?

Which I/O technique is better for real-time systems and why?

Conclusion

The Real-Time System exam paper evaluates knowledge at three levels. Section A focuses on basic definitions and concepts related to real-time computing systems. Section B examines scheduling algorithms and system design issues. Section C explores advanced concepts such as task classification, clock synchronization, and I/O operations. 

Understanding these concepts is essential for designing systems that must operate within strict timing constraints.