Query Optimization: Cost-Based Optimization |
Distributed query processing is fast becoming a reality.With the new emerging applications such as the grid applications, distributeddata processing becomes a complex undertaking due to the changes coming fromboth underlying networks and the requirements of grid-enabled databases. Recentdatabase research has demonstrated that memory access is more and more becominga significant—if not the major—cost component of database operations In this article, wepropose a generic technique to create accurate cost functions for databaseoperations. The method of optimizing the query by choosing a strategy thatresults in minimum cost i.e. Cost based Query Optimization. We calculate thecost of executing the different alternatives .The cost of executing a queryinclude the following components: secondary storage access cost, storage cost,computation cost, memory usage cost, communication cost. We identify a fewbasic memory access patterns and provide cost functions that estimate theiraccess costs for each level of the memory hierarchy. The cost functions areparameterized to accommodate various hardware characteristics appropriately.Database processes queries, Processing Selection Queries, Processing ProjectionQueries and Eliminating Duplicates, Processing Join Queries: Two plans have thesame cost through Improvement - block nested loops join, indexed nested loopsjoin, sort-merge join, hash join, Query Plans and Query Optimization forComplex Relational Expression.