Minimizing Waiting Time for Service Composition: A Frictional Approach

Dynamic Service composition is the art of composing a composite application from a set of service offerings from various providers at run time. Service composition is a complex task and the problem is with the construction of hierarchies of service providers where each service provider processes only a part of the application. This is further exacerbated by the fact that for every part of the application there are several potential service providers. Choosing the most appropriate service from functionally equivalent services is the issue being addressed here. In this paper, we focus on the waiting time of a service request as the criterion for selection. We present a technique borrowed and customized from the concept of friction in physics to select the best potential candidate service from a set of functionally equivalent ones, so that the waiting time for each service request is minimized. The technique not only facilitates the selection of the most appropriate service but also minimizes the waiting time of service requests. The model is validated through simulation and is shown to be able to equitably balance the load among similar service offering nodes.