Parallel Application Research Of Comprehensive Evaluation Decision-making Model In Cluster

As calculation brings more and more big effect in study of science and actual application, people have taken numerical value simulation as the basis of many decision-making. Parallel computing has developed at full speed since the nineties duo to the improvement of single processing machine speed, the maturity of parallel computer system structure, the normalization of data transmission network and lifting of wide transmission rate. Parallel computing has been paid close attention now and becomes a hot spot of international academic circles studies.This paper mainly studies the algorithm of comprehensive evaluation investment decision model and realizes with parallel, main content includes:(1)TOPSIS is technique for order preference by similarity to ideal solution, it orders in order to measure some solution near ideal solution and far away negative ideal solution by defining one degree in target space. Finally it respectively calculates the distance between every object and ideal solution and negative ideal solution, gaining every appraise scheme and optimum scheme approaching degree to order as the basis of valuation superior or inferior. This paper serially realizes three-dimensional TOPSIS and realizes parallel algorithm to comprehensive evaluation.Experiment testifies that parallel algorithm brings convenient and fast advantage for comprehensive evaluation with mass data.(2) Main characteristic of T.cover optimum investment decision-making model is handled problem evaluating many object and different index in some time and several valid periods. investment avails vector is random which indicates unit capital investment. Tactics goodness and badness doesn't depend on once investment, but considers m times general effect with investment. Therefore calculates optimum investment decision-making vector by folding up and optimum proportion through order. This paper serially realizes two-dimensional T.cover and realizes parallel algorithm. Experiment verifies feasibility and correctness, shows that the speedup and parallel efficiency is high enough and provides a new thought and method for economic effect comprehensive evaluation.(3)Entropy weight coefficient method according to character and concept quantizes information between every choosing scheme and subjective information and builds comprehensive evaluation decision-making model. It characteristic is comparing every appraising index to comparatively ideal value, which is an ideal evaluation model. This paper serially realizes model of Entropy weight coefficient method and realizes parallel algorithm. It not only resolves the application problem in the field of economic effect, but puts it further into extend education evaluation. Experiment result proves feasibility and validity of parallel algorithms, guarantes education evaluation to be fair, fairly carried on, takes this as the basis to science inference and prompts objective law in education phenomenon.