User-defined aggregate functions in database systems without native support

摘要

A computer-implemented method of using a database management system (DBMS) for providing an aggregate of a plurality of entities, constructing an aggregate of said plurality of entities in a memory to provide a result, and returning the result from the memory. Constructing the aggregate in memory includes storing in the memory a plurality of intermediate aggregation results each associated with a unique identifier. The unique identifier for each intermediate result is stored in the DBMS. Each unique identifier indicates information identifying a position within a sequence of generating the intermediate result and a pointer to a location in the memory where that intermediate result is stored.