Semantical partitioning of data

• US 7,814,045 B2
• Filed: 10/04/2006
• Issued: 10/12/2010
• Est. Priority Date: 10/04/2006
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: SAP AG (SAP SE)
• Inventors: Holtke, Rainer, Hermann, Alexander, Uhle, Michael Te, Schulte, Uwe
• Primary Examiner(s): Kim; Charles
• Assistant Examiner(s): MUELLER, KURT A
• Application Number:
  US11/543,328
• Publication Number:
  US 20080086478A1
• Time in Patent Office: 1,469 Days
• Field of Search: 707/600
• US Class Current: 707/600
• CPC Class Codes: G06F 16/254 Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses

Abstract

Data can be extracted from one or more data sources. Thereafter, the extracted data can be transformed to a format compatible with a plurality of structurally identical data targets so that the transformed data can be loaded into the plurality of data targets via data transfer processes. The data transfer processes define disjoint data sets among the plurality of data targets. Related apparatus, systems, methods, and articles are also described.

Claims

1. An article for semantic partitioning data stored in a data warehouse environment, the article comprising a tangible machine-readable storage medium embodying instructions that when performed by one or more machines result in operations comprising:

extracting, in a data warehousing environment, data from one or more data sources;

transforming the extracted data to a format compatible with a plurality of structurally identical data targets; and

loading, in parallel, the transformed data into the plurality of data targets via data transfer processes;

wherein;

the data targets comprises metaobjects in a database that is uniformly seen within a query definition and having data that is reported uniformly by one or more reporting modules,

the data transfer processes define disjoint data sets among the plurality of data targets so that each data target includes a distinct data set,

the transforming comprises automatically mapping data from the one or more data sources to data in the one or more data targets using predefined transformation rules,

the transformation rules map any number of source fields to at least one target field, wherein routines are used to implement complex transformation rules,

the routines are available as a rule type,

the rule type is a specific operation applied to relevant fields using a transformation rule;

at least one of the reporting modules is a MultiProvider;

at least one of the reporting modules is an Infocube

at least one of the data targets is an InfoProvider with data storage.

2. The article as in claim 1, wherein the loading comprises:

concurrently loading the transformed data into at least two of the data targets.

3. The article as in claim 1, wherein the extracted data is aggregated in the MultiProvider to enable the extracted data to be reported.

4. The article as in claim 3, wherein the article is further operable to cause data processing apparatus to perform operations comprising:

generating a query on top of the MultiProvider.

5. The article as in claim 1, wherein the extracted data is populated in the InfoCube which comprises a plurality of relational tables arranged according to a star schema.

6. An article for semantically partitioning data stored in a data warehouse, the article comprising a tangible machine-readable storage medium embodying instructions that when performed by one or more machines result in operations comprising:

deriving a plurality of structurally identical data targets from a master object, the data targets comprising metaobjects in a database that is uniformly seen as data providers within a query definition and having data that is reported uniformly through one or more reporting modules; and

defining disjoint data transfer processes coupling the data targets to one or more data sources, the data transfer processes defining disjoint semantic partitions among the plurality of structurally identical data targets so that each data target includes a distinct data set;

wherein at least one of the reporting modules is a MultiProvider;

wherein at least one of the reporting modules is an Infocube;

wherein at least one of the data targets is an InfoProvider with data storage.

7. The article as in claim 6, wherein the article is further operable to cause data processing apparatus to perform operations comprising:

defining an overflow data provider containing all data sets that are not associated with one or more of the data targets; and

defining an overflow data transfer process coupling the overflow data provider to the one or more data sources.

8. A semantically partitioned data warehousing system comprising:

one or more data sources;

a first data collection module characterizing data available in the one or more data sources, the first data collection module being connected to the one or more data sources via a first set of transformations, the first set of transformations mapping data from the one or more data sources from a first format to a second format;

a second data collection module characterizing data available for the plurality of data targets, the second data collection module being connected to the first data collection module via a second transformation;

two or more structurally identical data targets, the data targets each being identically connected to the second data collection module via a third set of transformations, the data targets being respectively coupled to the one or more data sources by disjoint data transfer processes so that each data target includes a distinct data set, the data targets comprising metaobjects in a database that is uniformly seen as data providers within a query definition and having data that is reported uniformly; and

one or more reporting modules to enable reporting the extracted data;

wherein;

at least one of the data collection modules comprises an InfoSource;

at least one of the reporting modules comprises a MultiProvider;

at least one of the data sources is an InfoProvider.

9. The system as in claim 8, wherein the disjoint data transfer processes are partitioned according to semantics, the semantics associating data sets derivable from the data sources with the data targets.

10. The system as in claim 8, further comprising:

a third data collection module characterizing data available in the one or more data targets.

11. The system as in claim 10, further comprising:

an overflow data provider characterizing all data sets that are not associated with one or more of the data targets, the overflow data provider being connected to the second data collection module.

12. The system as in claim 8, wherein the data targets are derived from a master meta data object.

13. The system as in claim 8, wherein the data targets hold disjunctive data sets.