Juanit Ablair


Big image

The same parts can be used in various combinations

A set of customers reaches a looking for a service, if this is not expected immediately, and leave the system once they have been served system. In some cases you can admit that customers leave the system if they get tired of waiting. The term customer queue management client is used in a general sense and does not mean to be a human being, can mean parts waiting their turn to be processed or a work list hoping to print to a printer in most red. Unique systems they can be represented as in Figure 1, it should be clear that a detailed representation requires defining numerous parameters and functions.

Queuing theory was originally a practical work. The first application of which the news is the Danish mathematician on telephone conversations in 1909, for calculating size switchboards. Then he became a theoretical concept that customer queue management got a great development, and for some years was once again talking about a concept applied but requires extensive analysis to convert the formulas into reality, or viceversa. En usual situations tail, Arrival is stochastic, ie the arrival depends on some random variable, in this case it is necessary to know the probability distribution between two successive arrivals' client.

In addition, it should be borne in mind if customers arrive independently or simultaneously. In the latter case (i.e., if they batch) would define the probability distribution thereof. It is also possible that clients are impatient. That is, they reach the customer queue management queue and if it is too long to leave, or after waiting a long time in the queue decide to leave.

Finally, it is possible that the pattern of arrival varies with time. If you kept constant we call stationary, if for example varies with the hour is. Lost time customer queue management servers may have a variable duty, in which case you have to associate, to define a probability function. They can also serve in batches or individually.

The service time may also vary with the number of customers in the queue, working faster or slower, and in this case is called dependent service patterns. As the pattern of arrivals pattern service may be varying over time. La queue discipline is the way in which customers are arranged in time to be served from the tail. When you think about queues it is accepted that the queue system solution normal queue discipline is FIFO (first meet who came first), however in many queues is common to use the LIFO discipline (attending first to last).

It is also possible to find prioritized sequence rules such as sequencing tasks with shorter or as first client types. In any case there are two general situations in which to work. In the first, called in English preemptive if a customer arrives in the queue with a superior customer order that is being dressed priority, this is removed giving way to the more important. Two new sub-cases appear: retired client has to start over, or customer returns where you left off. The second situation is the non-preemptive where the customer called higher priority waits for customer queue management runs being.
Big image
IBM Websphere MQ remote Queue Manager Configuration