Real-time

An operation within a larger dynamic system is called a real-time operation if the combined reaction- and operation-time of a task is shorter than the maximum delay that is allowed, in view of circumstances outside the operation. The task must also occur before the system to be controlled becomes unstable. A real-time operation is not necessarily fast, as slow systems can allow slow real-time operations. This applies for all types of dynamically changing systems. The polar opposite of a real-time operation is a batch job with interactive timesharing falling somewhere in-between the two extremes.

Alternately, a system is said to be hard real-time if the correctness of an operation depends not only upon the logical correctness of the operation but also upon the time at which it is performed. An operation performed after the deadline is, by definition, incorrect, and usually has no value. In a soft real-time system the value of an operation declines steadily after the deadline expires.

A typical example could be a computer-controlled braking system in a car. If the driver can stop a car before it hits a wall, the operation was in real-time; if the car hits the wall it was not. Many machines require real-time controllers to avoid "instability", which could lead to the accidental damage or destruction of the system, people, or objects.

Some real-time systems do not have such a constraint on delay as long as input data can be processed rapidly enough so that no backlog occurs. In a real-time Digital signal processing (DSP) system, the analyzed (input) and/or generated (output) samples (whether they are grouped together in large segments or processed individually) can be processed (or generated) continuously in the time it takes to input and/or output the same set of samples independent of the processing delay. Consider an audio DSP example: if a process requires 2.01 seconds to analyze or process 2.00 seconds of sound, it is not real-time. If it takes 1.99 seconds, it is (or can be made into) a real-time DSP process.

A common life example is that of standing in a queue (line of customers) waiting for the checkout in a grocery store. The queue is "real-time" if, on average, customers are being processed and outputted as quickly as they arrive. Under these circumstances, a queue of customers may occasionally form, but the queue can't grow without bounds.

If, on the other hand, customers are not processed as quickly as they arrive, then the queue of waiting customers will grow without limits. This system is not real-time and would definitely be deemed a failure.

A more subtle failure might occur even if the system can, on average, process customers as quickly as they arrive. A burst of arrivals or some slow processing of the existing customers will cause the queue to fill. If the queue becomes sufficiently long, customers will be discouraged and leave without making a purchase. This system would also have to be deemed a failure, even though it meets the strict definition of real time as described above. In this case, the system has a limited "buffer capacity" (tolerable queue length) and it must process customers fast enough to keep the queue shorter than that tolerable queue length even under a worst-case load of arriving customers and the slowest processing of existing customers.

Real stores (and real real-time systems) often need better than real-time performance, and worst-case scenarios must be carefully evaluated! In the case of the hypothetical grocery store, they may define a given "service level", expect real-time processing up to that service level, and accept a certain loss of customers beyond that service level. A missile designer, on the other hand, may not be willing to define a service level beyond which failures can be expected; for the missile designer, no failure is acceptable.

In the economy, real-time systems are information technologies, which provide real-time access to information or data. The ability of a company to process its data in real time increases the competitiveness of the company. Real-time processing systems are new technologies and will improve during the next decades. Gartner forecasts a fast increase and use of these real-time systems.

See also

• Queueing theory
• Rate-monotonic scheduling
• Real-time computing
• Real-time control
• Real-time operating system
• Real-time video editing
• Embedded system
• Frame rate
• Real-time strategy gamede:Echtzeit

es:Informática o Computación en tiempo real fr:Temps réel nl:Real-time pl:Czas rzeczywisty