PROJECT TITLE :

Fast Multi-Criteria Service Selection for Multi-User Composite Applications

ABSTRACT:

Paradigms such as Software as a Service (SaaS) and Service-Based Systems (SBSs), which are becoming more prevalent as cloud computing becomes the dominant facet of software engineering, are coming into focus. They are built on cloud services that can respond to multiple client requests at the same time. When there are requirements for multiple Quality of Service (QoS) attributes that need to be satisfied for a large number of users at the same time, selecting the actual service instance for the request can be a challenge. When we take into account the composite nature of user applications, which can consist of a large number of tasks and require that quality of service (QoS) properties be calculated over the entire composition, the problem becomes more complicated. The currently available solutions to this issue are neither effective nor applicable in a wide variety of contexts. In this article, we propose a fast heuristic method for multi-criteria service selection. This method is intended for multi-user composite workflows, and its primary objective is to satisfy all of the given QoS requirements, or at least as many of them as is humanly possible. Using a global-aware utility cost that is based on expected compositional QoS, as well as iterative solution improvements, the proposed method breaks the problem down into several independent transportation problems. The proposed method, in addition to being more general than the existing approaches, turns out to be more efficient than the alternatives (up to 5 times faster), as was demonstrated by extensive experiments that covered both specific and more general cases.