QoS correlation-based service composition algorithm for multi-constraint optimal path selection

As network services tend to be integrated to provide a better quality of service (QoS) to customers, correlations appear to be useful measurements to better design service compositions of an integrated network. However, as the integration intensifies, the determination of the service compositions of the network requires the ease of computational issues in this dynamic environment, which are resolved by the cloud computing platforms. The manuscript proposes an algorithm based on multi-constraint optimal path selection (MCOPS) that benefits from available notions such as QoS correlation criteria and correlation ratios, and skyline algorithm to construct a novel directed cyclic graph whose calculations are conducted on the cloud platform dynamically. Both cost and delay attributes in the construction of network service compositions for customers are included in the graph. Simulations suggest that both average calculation time and the quality of the path solution are substantially enhanced with the utilization of cloud computing in network service compositions. Consequently, a better service composition plan (SCP) is attained when a correlated structure is assumed to exist.