Abstract
With the proliferation of Software-as-a-Service (SaaS) in the cloud environment, it is difficult for users to search for the right service that satisfies all their needs. In addition, services may provide the same functionality but differ in their characteristics or Quality of Service attributes (QoS). In this chapter, we present a comprehensive survey on cloud services discovery and selection research approaches. Based on this survey, a complete system with efficient service description model, discovery, and selection mechanisms is urgently required. Therefore, we propose a semantic-based SaaS publication, discovery, and selection system, which assists the user in finding and choosing the best SaaS service that meets his functional and non-functional requirements. The basic building block of the proposed system is the unified ontology, which combines services domain knowledge, SaaS characteristics, QoS metrics, and real SaaS offers. A hybrid service matchmaking algorithm is introduced based on the proposed unified ontology. It integrates semantic-based meta data and ontology-based matching. Ontology-based matching integrates distance-based and content-based concept similarity measures. The matchmaking algorithm is used in clustering the SaaS offers into functional groups to speed up the matching process. In the selection process, the discovered services are filtered based on their characteristics, and then they are ranked based on their QoS attributes. Case studies, prototypical implementation results, and evaluation are presented to demonstrate the effectiveness of the proposed system With the proliferation of Software-as-a-Service (SaaS) in the cloud environment, it is difficult for users to search for the right service that satisfies all their needs. In addition, services may provide the same functionality but differ in their characteristics or Quality of Service attributes (QoS). In this chapter, we present a comprehensive survey on cloud services discovery and selection research approaches. Based on survey, a complete system with efficient service description model, discovery, and selection mechanisms is urgently required to assist the user in finding and choosing the best SaaS service that meets his functional and non-functional requirements. Therefore, we propose a semantic-based SaaS publication, discovery, and selection system. We developed a unified ontology that combines services domain knowledge, SaaS characteristics, QoS metrics, and real SaaS offers. A hybrid service matchmaking algorithm is introduced based on the proposed unified ontology. It integrates semantic-based metadata and ontology-based matching. Ontological similarity integrates distance-based and content-based concept similarity measures. It is used in clustering the SaaS offers into functional groups to speed up the matching process. Case studies, prototypical implementation results, and evaluation are presented to demonstrate the effectiveness of the proposed system.