S-ABC - A Service-Oriented Artificial Bee Colony Algorithm for Global Optimal Services Selection in Concurrent Requests Environment

In the Big Data environment, more and more services with identical functionalities are emerging on the Internet, but the Quality of Service (QoS) of these services are different, therefore, service selection based on QoS becomes a challenge problem. In the past few years, extensive studies have been carried out on this problem and several approaches are proposed. However, existing research work mainly focuses on service selection for single request. In fact, with the rapid growth of service users, concurrent requests for the same service are taking place frequently. How to find the global optimal services selection scheme for concurrent requests so as to improve users' satisfaction and service broker's revenue has become a difficult problem. For this issue, this paper firstly transforms the problem of global optimal services selection for concurrent requests to a combinatorial optimization problem, then, it proposes a new evaluation model including admission control operator, users' satisfaction and service broker's revenue evaluation operators, finally, based on two important service domain characteristics(Similarity and Priori), a new Service-Oriented Artificial Bee Colony algorithm (S-ABC) is constructed for solving the global optimal services selection problem. Experimental results show that proposed approach in this paper is effective.