DYNAMIC SELECTION OF SOURCE TABLE FOR DB ROLLUP AGGREGATION AND QUERY REWRITE BASED ON MODEL DRIVEN DEFINITIONS AND CARDINALITY ESTIMATES

摘要

A method for optimizing data access in a data warehouse having a model driven architecture is provided. A data model for storing time series measurement data representing a plurality of dimensions and measurements is received. The data model includes one or more aggregation tables. A query having one or more predicates is received from a user. The one or more predicates are analyzed to determine whether the received query can be optimized using the existing aggregation tables. The received query is optimized using the aggregation tables, in response to determining that such query optimization is possible.

主权项

1. A method for optimizing data access in a data warehouse having a model driven architecture, the method comprising:
receiving a data model for storing time series measurement data representing a plurality of dimensions and measurements, the data model having one or more aggregation tables; receiving a query from a user, the query having one or more predicates; analyzing, by a processor, the one or more predicates to determine whether the received query can be optimized using the one or more aggregation tables; and optimizing, by the processor, the received query using the one or more aggregation tables, in response to determining that the received query can be optimized.