OLAP grouping aggregation method based on function dependency relationship

Abstract

The invention relates to an OLAP grouping aggregation method based on a function dependency relationship. The method comprises following steps: a function dependency tree structure among grouping attributes is defined, and the function dependency relationship between the grouping attributes and attributes among tables is defined; the grouping attribute of the GROUP BY clause is detected according to the grouping attribute function dependency tree; a query grouping attribute is divided into two subgroups which are formed by function dependency key attributes and other grouping attributes respectively; mode resolution is performed on a dimension table according to the function dependency relationship, wherein the dimension table adopts surrogate key mechanism and memory column storage mechanism; under the mechanisms of dimension table mode resolution and surrogate key index, grouping aggregation calculation further converts the subgroup attributes into function dependency attributes; after the grouping aggregation calculation is completed, the function dependency attributes are directly mapped to a memory offset address of related dimension table records according to function dependency attribute values to complete the operation of extracting other related grouping attributes. By means of the method, the size of grouping attributes can be reduced, and the Hash grouping calculating efficiency is increased.

Description

technical field [0001] The invention relates to an OLAP grouping aggregation method, in particular to an OLAP grouping aggregation method which reduces the grouping attributes through the functional dependency relationship between the same or different table attributes in a database, and optimizes the grouping aggregation computing performance based on the functional dependency relationship. Background technique [0002] At present, grouping and aggregation operations are an important function of OLAP queries. In relational databases, records are grouped and aggregated mainly through hash tables. In OLAP queries, the GROUP BY attribute is a grouping attribute. When there are many grouping attributes or the grouping attributes involve longer character attributes, the database directly uses the longer attribute as the hash key value for group calculation. When hash mapping A higher CPU calculation cost is generated, and the group hash table needs to store a longer key value, w...

AI Technical Summary

Problems solved by technology

In OLAP queries, the GROUP BY attribute is a grouping attribute. When there are many grouping attributes or the grouping attributes involve longer character attributes, the database directly uses the longer attribute as the hash key value for group calculation. When hash mapping A higher CPU calculation cost is generated, and the group hash table needs to store a longer key value, which leads to an increase in the space of the hash table, more cache misses, and increases the memory access delay of the group calculation

Images