PROJECT TITLE :

Cognitive Storage for Big Data

ABSTRACT:

Stochastic Trust-Region Response-Surface methodology (STRONG) may be a new response-surface-primarily based framework for simulation optimization. The appeal of STRONG lies in that it preserves the advantages, nevertheless eliminates the disadvantages, of traditional response surface methodology (RSM) that has been used for more than fifty years. Specifically, STRONG does not require human involvement in the search method and will guarantee to converge to the true optimum with chance one (w.p.1). During this paper, we have a tendency to propose an improved framework, known as STRONG-X, that enhances the efficiency and efficacy of STRONG to widen its applicability to a lot of practical issues. For efficiency improvement, STRONG-X includes a newly-developed experimental scheme that consists of construction of optimal simulation styles and an assignment strategy for random range streams to get computational gains. For efficacy improvement, a new variant, referred to as STRONG-XG, is developed to realize convergence beneath typically-distributed responses, as opposed to STRONG and STRONG-X where convergence is guaranteed only when the response is traditional. An intensive numerical study is conducted to evaluate the efficiency and efficacy of STRONG-X and STRONG-XG. Moreover, 2 illustrative examples are provided to point out the viability of STRONG-X and STRONG-XG in sensible settings.