Analysis of Cost Estimation used in Query Optimization of Fuzzy Relational Databases based on Sort-Merge Algorithm

摘要

Query optimization in fuzzy relational databases aims to come out with a minimal execution cost for the available execution strategies. Each type of cost is estimated by a cost function. All cost functions together with their parameters and assumptions forms a cost model for the fuzzy query optimizer. The cost function usually takes the size of tables as inputs. It is possible that exact information is not available where by fuzzy data is assumed. Also estimating the nature of different cost models needs to be examined. A fuzzy cost function is used for this purpose which produces a fuzzy value that represents a soft estimate of the real cost while a traditional crisp cost function, produces a hard crisp estimate.