A Constraint-based Mining Approach for Multi-attribute Index Selection

The index selection problem (ISP) concerns the selection of an appropriate indexes set to minimize the total cost for a given workload under storage constraint. Since the ISP has been proven to be an NP-hard problem, most studies focus on heuristic algorithms to obtain approximate solutions. The problem becomes more difficult for indexes defined on multiple tables such as bitmap join indexes, since it requires the exploration of a large search space. Studies dealing with the problem of selecting bitmap join indexes mainly focused on proposing pruning solutions of the search space by the means of data mining techniques or heuristic strategies. The main shortcoming of these approaches is that the indexes selection process is performed in two steps. The generation of a large number of indexes is followed by a pruning phase. An alternative is to constrain the input data earlier in the selection process thereby reducing the output size to directly discover indexes that are of interest for the administrator. For example, to select a set of indexes, the administrator may put limits on the number of attributes or the cardinality of the attributes to be included in the indexes configuration he is seeking. In this paper we addressed the bitmap join indexes selection problem using a constraint-based approach. Unlike previous approaches, the selection is performed in one step by introducing constraints in the selection process. The proposed approach is evaluated using APB-1 benchmark.