SLA_oriented service selection in cloud environment: A PROMETHEE_based approach

In an open cloud computing environment, many functional-equivalent services can be provided by different providers with different SLA (service Level agreement) levels, then it is an important problem for service consumers to select the optimal one from lots of service candidates. Considering drawbacks of existing methods, we propose a systematic approach for this multi-objective optimization problem. First, we present a novel two-stage ranking model, in which the PROMETHEE is introduced on the basis of classical Pareto-optimization method, to provide further ranking and selection among Pareto optimal candidates. Second, we present a BNL(block nested loop) algorithm according to the proposed two-stage ranking model, which can efficiently find and return top-k candidates for the service selection. Finally, we evaluate the effectiveness and efficiency of our algorithm through a set of experimental studies.