Scalable indexing architecture
AssigneeQliktech International AB
InventorsJosé Diaz López, Johan Nilsson, Khoa Tan Nguyen, Quan Sun, Vladimir Mordvinov, Karl Tracy Blomquist, Pawel Rafal Krupinski
Life of this patent
Estimating the term…
Abstract
Provided are methods and systems for improved data management, data analysis, and processing. A computing device may receive one or more global queries. Each of the global queries may be associated with one or more local components, such as individual nodes of a cloud network. The computing device may determine one or more indexlets associated with each local component. The computing device may then generate query results for the one or more global queries using a global query plan and the one or more indexlets.
Claims20
A method comprising:
- receiving appended data for a data model comprising a plurality of tables, wherein the data model is associated with a hypercube that is generated based on the plurality of tables excluding the appended data;
- causing, based on the appended data, a plurality of indexlet indexers to generate a plurality of indexlets for the appended data, wherein each indexlet of the plurality of indexlets comprises a bidirectional index for at least one partition of at least one table, of the plurality of tables, associated with the appended data;
- determining, based on one or more global queries associated with the data model, that the plurality of indexlets are associated with the one or more global queries;
- and
- generating, based on the one or more global queries and each bidirectional index of each indexlet of the plurality of indexlets, partial query results for the one or more global queries, wherein the partial query results are generated without regenerating the hypercube.
A non-transitory computer-readable storage medium comprising computer-executable instructions that, when executed by a computing device, cause the computing device to:
- receive appended data for a data model comprising a plurality of tables, wherein the data model is associated with a hypercube that is generated based on the plurality of tables excluding the appended data;
- cause, based on the appended data, a plurality of indexlet indexers to generate a plurality of indexlets for the appended data, wherein each indexlet of the plurality of indexlets comprises a bidirectional index for at least one partition of at least one table, of the plurality of tables, associated with the appended data;
- determine, based on one or more global queries associated with the data model, that the plurality of indexlets are associated with the one or more global queries;
- and
- generate, based on the one or more global queries and each bidirectional index of each indexlet of the plurality of indexlets, partial query results for the one or more global queries, wherein the partial query results are generated without regenerating the hypercube.
An apparatus comprising:
- one or more processors;
- and
- a memory comprising processor-executable instructions that, when executed by the one or more processors, cause the apparatus to:
- receive appended data for a data model comprising a plurality of tables, wherein the data model is associated with a hypercube that is generated based on the plurality of tables excluding the appended data;
- cause, based on the appended data, a plurality of indexlet indexers to generate a plurality of indexlets for the appended data, wherein each indexlet of the plurality of indexlets comprises a bidirectional index for at least one partition of at least one table, of the plurality of tables, associated with the appended data;
- determine, based on one or more global queries associated with the data model, that the plurality of indexlets are associated with the one or more global queries;
- and
- generate, based on the one or more global queries and each bidirectional index of each indexlet of the plurality of indexlets, partial query results for the one or more global queries, wherein the partial query results are generated without regenerating the hypercube.
Dependencies are inferred from the claim text.