Abstract: Methods and systems for using algorithms in relational database management systems (RDBMSs) for incremental maintenance of materialized views with outerjoins are disclosed. The algorithms achieve the following goals with respect to a class of materialized outerjoin views and the performance of update operations: relax the requirement for the existence of the primary key attributes in a select list of the view to only some of the relations (i.e., the relations referenced as a preserved side in an outerjoin); relax null-intolerant property requirements for some predicates used in the view definition (i.e., predicates referencing relations which can be null-supplied by more than one outerjoin); and implement maintenance of outerjoin views by using one update statement (e.g., MERGE, UPDATE, INSERT, or DELETE) per view for each relation referenced in the view. The algorithms allow design and implementation of the incremental maintenance of materialized views with outerjoins to be integrated into an RDBMS.

Abstract: In an optimizer within a Relational Database Management System improved ways in which a search space (the universe of possible join combinations) is generated and managed and improved ways in which the elements of a search space are evaluated so that among other things unpromising elements are efficiently dropped (pruned).

Abstract: Systems, methods and articles of manufacture for determining at least one algorithm from a pool of algorithms to optimize a query. An embodiment includes building optimization blocks for the subplan, determining the complexity of the subplan based on the optimization blocks and selecting an algorithm from a pool of search space generation algorithms based on complexity of the subplan and characteristics of the selected algorithm, and optimizing the subplan with the selected algorithm.

Abstract: A system, method, and computer program product are provided for generating a graphical representation of a query optimization process. The method comprises the steps of parsing a search space log, presenting one or more evaluated access plans on an axis of a timeline, identifying a best access plan on the timeline, and outputting a graphical representation of the timeline. An additional system, method, and computer program product are provided for recording a query optimization process of a query optimizer.

Abstract: Methods and systems for using algorithms in relational database management systems (RDBMSs) for incremental maintenance of materialized views with outerjoins are disclosed. The algorithms achieve the following goals with respect to a class of materialized outerjoin views and the performance of update operations: relax the requirement for the existence of the primary key attributes in a select list of the view to only some of the relations (i.e., the relations referenced as a preserved side in an outerjoin); relax null-intolerant property requirements for some predicates used in the view definition (i.e., predicates referencing relations which can be null-supplied by more than one outerjoin); and implement maintenance of outerjoin views by using one update statement (e.g., MERGE, UPDATE, INSERT, or DELETE) per view for each relation referenced in the view. The algorithms allow design and implementation of the incremental maintenance of materialized views with outerjoins to be integrated into an RDBMS.

Abstract: A system, method, and computer program product are provided for immediately updating a materialized view in a relational database management system. The method includes the steps of creating a materialized view defined over a base table, executing an update statement on the base table, computing a delta of the base table by comparing images of the base table before execution of the update statement and after execution of the update statement, computing a delta of the materialized view based on the delta of the base table, and updating the materialized view using the delta of the materialized view.

Abstract: A system, method, and computer program product are provided for generating a graphical representation of a query optimization process. The method comprises the steps of parsing a search space log, presenting one or more evaluated access plans on an axis of a timeline, identifying a best access plan on the timeline, and outputting a graphical representation of the timeline. An additional system, method, and computer program product are provided for recording a query optimization process of a query optimizer.

Abstract: A system, method, and computer program product are provided for immediately updating a materialized view in a relational database management system. The method includes the steps of creating a materialized view defined over a base table, executing an update statement on the base table, computing a delta of the base table by comparing images of the base table before execution of the update statement and after execution of the update statement, computing a delta of the materialized view based on the delta of the base table, and updating the materialized view using the delta of the materialized view.

Abstract: A system providing methodology for cost-based enumeration of subqueries using a left-deep tree join enumeration algorithm is described.

Abstract: Systems, methods and computer program products for optimization of query evaluation using materialized views in a database management system (DBMS) are described herein. During a pre-optimization phase, building blocks are generated that can be used to generate valid partial execution plans, where at least one of the building blocks can be a materialized view. During an enumeration phase, a plurality of execution plans for the query are generated using the building blocks. Some of the execution plans may include materialized views.

Abstract: Techniques for left outer join (LOJ) and right outer join (ROJ) optimization are described herein. Each LOJ/ROJ is an expression that references a preserved side and a null supplying side. If Condition 1 is true, and either Condition 2 or Condition 3 is true, then the LOJ/ROJ is evaluated by evaluating the expression referencing only the preserved side, and not the null supplying side. Conditions 1, 2 and 3 are defined as follows: Condition 1: An expression from the null supplying side is not used outside of the LOJ/ROJ. Condition 2: The LOJ/ROJ is required to return only distinct values from the preserved side. Condition 3: The null supplying side returns at most one row for each row returned by the preserved side.

Abstract: A database system with methodology for automated determination and selection of optimal indexes is described. In one embodiment, for example, in a database system, a method of the present invention is described for recommending database indexes to be created for optimizing system performance, the method comprises steps of: capturing a workload representative of database queries employed during system use; creating virtual indexes for optimizing system performance during execution of the database queries captured in the workload; computing cost benefits for different combinations of the virtual indexes; and recommending physical indexes to be created based on virtual indexes that have favorable cost benefits for the captured workload.

Abstract: Systems, methods and computer program products for optimization of query evaluation using materialized views in a database management system (DBMS) are described herein. During a pre-optimization phase, building blocks are generated that can be used to generate valid partial execution plans, where at least one of the building blocks can be a materialized view. During an enumeration phase, a plurality of execution plans for the query are generated using the building blocks. Some of the execution plans may include materialized views.

Abstract: Techniques for left outer join (LOJ) and right outer join (ROJ) optimization are described herein. Each LOJ/ROJ is an expression that references a preserved side and a null supplying side. If Condition 1 is true, and either Condition 2 or Condition 3 is true, then the LOJ/ROJ is evaluated by evaluating the expression referencing only the preserved side, and not the null supplying side. Conditions 1, 2 and 3 are defined as follows: Condition 1: An expression from the null supplying side is not used outside of the LOJ/ROJ. Condition 2: The LOJ/ROJ is required to return only distinct values from the preserved side. Condition 3: The null supplying side returns at most one row for each row returned by the preserved side.

Abstract: A database system and methodology for generalized order optimization is described.

Abstract: A system and methodology for generating bushy trees using a left-deep tree join enumeration strategy for optimizing execution of a database query is described. In response to receipt of a query specifying at least one join condition between two or more database tables, each query block comprising an atomic portion of the query is identified and subplans are created for each query block based on grouping portions of each query block. At least one favorable access plan is determined for each subplan of each query block based, at least in part, on estimated execution costs. An optimal access plan for each query block is generated based upon at least one favorable access plan determined for each subplan. A query execution plan is then constructed based upon the optimal access plan generated for each query block.

Abstract: A database system and methodology for generalized order optimization is described.

Abstract: A database system with methodology for automated determination and selection of optimal indexes is described. In one embodiment, for example, in a database system, a method of the present invention is described for recommending database indexes to be created for optimizing system performance, the method comprises steps of: capturing a workload representative of database queries employed during system use; creating virtual indexes for optimizing system performance during execution of the database queries captured in the workload; computing cost benefits for different combinations of the virtual indexes; and recommending physical indexes to be created based on virtual indexes that have favorable cost benefits for the captured workload.

Abstract: A system providing methodology for cost-based enumeration of subqueries using a left-deep tree join enumeration algorithm is described.

Abstract: A system and methodology for generating bushy trees using a left-deep tree join enumeration strategy for optimizing execution of a database query is described. In response to receipt of a query specifying at least one join condition between two or more database tables, each query block comprising an atomic portion of the query is identified and subplans are created for each query block based on grouping portions of each query block. At least one favorable access plan is determined for each subplan of each query block based, at least in part, on estimated execution costs. An optimal access plan for each query block is generated based upon at least one favorable access plan determined for each subplan. A query execution plan is then constructed based upon the optimal access plan generated for each query block.