The model is fed with measurements of transaction resource demands (CPU, disk I/O, LAN, WAN), weighted by the transaction-mix (business transactions per unit of time). The weighted transaction resource demands are added-up to obtain the resource demands and divided by the resource capacity to obtain the resource loads. Changes in response time can also be predicted by the model. For example, in a simple case with a single resource, the response time formula: R=S/(1-U) where R=response_time, S=service_time, U=utilization, will calculate the response time as the utilization of that resource varies between 0=0% busy to 1=100% busy.2 This formula is a good approximation of the more complex math of queueing theory and requires adjustment when dealing with multiple resources.
Wescott, Bob (2013). The Every Computer Performance Book, Chapter 7: Modeling. CreateSpace. ISBN 978-1482657753. 978-1482657753 ↩
Wescott, Bob (2013). The Every Computer Performance Book, Chapter 3: Useful laws. CreateSpace. ISBN 978-1482657753. 978-1482657753 ↩