The use of SQO rules in DMLs

Abstract

Semantic query optimisation (SQO) is a query optimisation approach and has been researched in [M. Hammer et al., 1975], especially by King who has announced it firstly and proved the incomes [J.J. King, 1981]. The approach uses learned-rules from the past queries in order to execute new queries more intelligently without the need of the database access. It has several components which are query representation, query optimisation, automatic rule derivation and rule maintenance. Although the approach has many advantages to have the past experiences in each executed-queries, the use of rules are only limited to be used for SELECT statement by many researchers. In this paper we focused on the use of SQO rules in data manipulation languages (DMLs) such as UPDATE, DELETE and INSERT in order to decrease the DML's execution time. In this paper, we have introduced how to use these rules in DMLs and then according to the features of the rules, we identify the performance differences for each DMLs.