PROJECT TITLE :

Resource renting for periodical cloud workflow applications - 2017

ABSTRACT:

Cloud computing may be a new resource provisioning mechanism, which represents a convenient approach for users to access totally different computing resources. Periodical workflow applications commonly exist in scientific and business analysis, among many other fields. One among the foremost difficult problems is to work out the proper amount of resources for multiple periodical workflow applications. In this paper, the periodical workflow applications scheduling downside with total renting price minimization is taken into account. The novelty of this work depends precisely on this objective function, which is additional realistic in follow than the additional commonly thought of makespan minimization. An integer programming model is constructed for the problem under study. A Precedence Tree primarily based Heuristic (PTH) is developed that considers 3 varieties of initial schedule construction strategies. Based mostly on the initial schedule, two improvement procedures are presented. The proposed strategies are compared with existing algorithms for the related makespan primarily based multiple workflow scheduling drawback. Experimental and statistical results demonstrate the effectiveness and efficiency of the proposed algorithm.