Advanced Real Time Systems

Advanced Real Time Systems

T.R. Gopalakrishnan Nair (Prince Mohammad bin Fahd University (PMU), Saudi Arabia & Dayananda Sagar Institutions (DSI), India)
DOI: 10.4018/978-1-4666-5888-2.ch689

Chapter Preview



Real-time system has emerged to be one of the major application domains in computer science and engineering, in today’s scenario. A real-time system is a hardware cum software unit that must process information and produce a response within a specified time, else risk severe consequences, including failure. Lui Sha (2004) defined it as a system with explicit deterministic or probabilistic timing requirements. For example in avionics, flight control software must complete its execution in a specified time in order to accurately control the flight. In health monitoring, the software to monitor the patient status in ICU must get fully executed within its specified interval. Various other examples can be identified in automated digital control, nuclear power plant, automobile, ABS and telecommunications systems.

In these systems, the response must be produced by computing unit at the specified time limit and it is called a deadline. Based on their deadline commitments, real time systems are classified into hard and soft real time systems. Hard real time system is a computer system in which the response occurs within a specified time. Failure to meet such a timing requirement can have catastrophic consequences. A system in which the missed deadline does not cause any catastrophe but only degrades performance is termed as soft real time system. The classification of these hard and soft real time systems depends on the application and not on the computer system. The correctness of the response produced depends much on the logical design of the system and also the time at which the response was produced. The real time systems may enter into an unstable state when the expected response is not produced at the right time or when there is a sudden change in the schedule of tasks because of external catastrophe. The advances in real-time system are needed to detect the catastrophe and reschedule the tasks to maintain the stability of the system.

Before discussing the advances in real-time system, it is better to understand the core concepts of a real-time system. Here, a Job is a unit of work which is properly scheduled and executed by the system. A task is a set of selected jobs. Each task is characterized through several parameters. They are temporal parameters, functional parameters, resource parameters and interconnection parameters. The temporal parameter deals with the timing constraints and behavior of a task. The functional parameter deals with the intrinsic properties of the task. Resource parameter deals with the resource requirements like processor, and memory. Finally, the interconnection parameter deals with the dependency among the tasks. The primary properties of any real time system can be listed as time constraints involved, embedded nature, safety-critical issues, concurrency of tasks, distributed and feedback structure, task criticality, custom hardware, reactive nature, stability of operation, and the ways of exception handling. A real time task is generated when some event occurs. So the number of task emanating in system is more or less directly proportional to the event occurrence. When more number of events occur, the more number of tasks get generated. For example,temperature sensing task occurs periodically in a hard RTS, which is expected to monitor and control a heating furnace.

An adaptive or intelligent real time system needs to have additional parameters to adjust or adapt to the changing environment. The conventional real time system is not capable of handling the multi criticality of tasks in unusual circumstances. However, the general schemes like earliest deadline first scheduling algorithm, and some other priority based algorithms can precisely schedule all the tasks and dispatch them to the processors. The online scheduling algorithms can use the schedulability test to check whether a newly arrived different criticality task can be allowed into the system. The basic properties of any real time task can be depicted through its parameters.

Each task occurring in the system can have the following parameters.

Key Terms in this Chapter

Execution Time: The amount of time taken required by task to complete its execution.

Deadline: The instant of time at which a task must complete its execution.

Release Time: The instant of time at which the task becomes available for execution.

Response Time: The length of time from the release time of the job to the time instant when it completes.

Complete Chapter List

Search this Book: