Cloud capacity planning based on simulation and genetic algorithms

Reducing spending on information technology is one important area that enable enterprises to reduce cost. One area where this can be done is to use cloud computing. Cloud computing key benefits include scalability, instant provisioning, virtualized resources, and cost effectiveness. There are different ways to deploy cloud resources such as public, private, and hybrid cloud. Business requirements determine the best deployment model to use. In this work, we have built a simulation model based on genetic programming to find the optimal combination of private and public cloud resources to satisfy a pattern of demand over the planning period as well as the optimal guaranteed service level. Our main findings is that the optimal level of private computing capacity depends to a large extent on the shape of the demand curve, negative exponential or normally for example. Variations in demand within the same family of demand distributions have a very small effect on capacity for the same mean demand over the planning period but significant impact on capacity utilization and cost. The distinguishing feature of our model is that it can handle any theoretical or an ad hoc demand probability distribution. In addition, our computational scheme allows for any random variation in any of the parameters affecting the total cost of cloud resources consumed as long as this variation can be described by an estimated parametric or empirical probability density function. In addition, the model can be easily modified to determine the optimal total cost with respect to any parameters that can be used as decision variables. The accuracy and correctness of the model was tested against results obtained from a mathematical model based on an exponential probability distribution with almost identical results.