A recursive pattern recognition approach to selection web services in cloud environment

Abstract

There are optimization problems in the Cloud for the selection of web services, due to the large number of services available by different cloud providers and the diversity of quality of service parameters of each of them. This work proposes the adaptation of a pattern recognition model based on the systematic functioning of the brain called Ar2p for the selection of web services in composition activities in Cloud environments. The web serice are represented as patterns to be recognized by Ar2p, which determines the necessary and sufficient web services that constitute the composition of services that meet its functional and non-functional requirements. The services composition and activity selection have been formalized through a logical-mathematical model of web service recognition mechanisms in two steps, one that describes the syntactic search of the service and the second, which offers filtering through quality of service parameters. An adaptive implementation of the final model allows its recognition modules to be provided with any desired optimization strategy.