Skip to Content

Service demand

Service demand is the time spent by a system resource processing a single business driver unit. It is the basis for building a Working-with-queuing-network-models of the system.

The business driver map analysis estimates and reports the confidence interval and service demand for each associated business driver or resource. The following table shows an example of the service demand for the two business drivers defined in the business driver map:

In this example:

  • Each page request consumes 10 -5 seconds of AS-CPU.
  • Each sent message consumes 2 x 10 -4 seconds of AS-CPU.

The overall application server CPU utilization percentage for these business drivers can be calculated as
((PageRequests/sec x 10 -5 sec) + (SentMessages/sec x (2 x 10 -4 sec))) * 100.

Even if page requests and sent messages have a hourly or daily resolution, the service demand is calculated and expressed in seconds.

Warning

Note

Mathematically, service demand is the slope of a linear model that expresses resource utilization as a function of the business driver value. If a resource is shared between different business drivers (multiple associations on the same business driver map line), the mathematical model becomes more complex. It then expresses resource utilization as a linear function of all business drivers simultaneously (multivariate model). This way, the business driver map is able to perform a multivariate regression analysis, based on the non-negative least squares technique, to estimate service demand values.

The following figure shows an example of how service demand for a performance versus load analysis appears in BMC Helix Capacity Optimization.

Business driver map of a performance versus load analysis – Service Demand tab
map3_SD.png

 

Tip: For faster searching, add an asterisk to the end of your partial query. Example: cert*

BMC Helix Capacity Optimization 19.11