|
|
|
|
|
|
|
Figure 6.16
The coefficient of variance (c2) is the key parameter in
describing the service distribution. |
|
|
|
|
|
|
|
|
General Service-Time Distribution |
|
|
|
|
|
|
|
|
Service time may be constant or variable. Distributions can be categorized by the behavior of c, the coefficient of variation for the service time. Now c2 is the squared coefficient of service time variance, or |
|
|
|
|
|
|
|
|
For constant service time, c2 = 0, while for the exponential service time distribution, c2 = 1. The factor c2 is the primary parameter used in describing the distribution of service times (Figure 6.16). |
|
|
|
|
|
|
|
|
We indicate the use of a particular probability distribution by the following abbreviations: |
|
|
|
|
| | | | | | | | | | |
|
|
|
|
c = arbitrary, defined by coef. of variance |
|
|
|
| | | | | | | | | |
|
|
|
|
|
|
where s is the standard deviation of the service time (Ts) distribution and 1/m is the mean service time, Ts. The occupancy r is defined as the ratio of the request rate to the service rate, r = l/m. |
|
|
|
|
|
|
|
|
Queue models are categorized by the triple: |
|
|
|
 |
|
|
|
|
Arrival Distribution/Service Distribution/Number of Servers. |
|
|
|
|
|
|
|
|
Thus, M/M/1 is a single-server queue with Poisson arrival and exponential service distributions. |
|
|
|
|
|
