System and method for analyzing and reporting extensible data from multiple sources in multiple formats
A system and method for analyzing and reporting data from multiple sources is provided. The system is a foundation for an analytical platform that covers not only traditional relational data, but also a new generation of extensible data formats designed for the web, such as those based on XML (FIXML, FpML, ebXML, XBRL, ACORD, etc.), as well as HTML, E-mail, Excel, PDF, and others. In a preferred embodiment, the eXtensible on-line analytical processing (XOLAP), is a scalable client/server platform that allows the multi-dimensional analysis of modern data types, as well as traditional relational data, by bringing them all into an internal common XML-based model, without the time and expense of creating a data warehouse.
1. An extended online analytical server, comprising:
one or more adaptors associated with one or more data sources, each adaptor converting one or more pieces of data from a particular data source having a particular data format into a common data format;
an aggregator that aggregates the one or more pieces of data from the one or more data sources to generate aggregated pieces of data;
a virtual schema generator that generates a virtual schema from the aggregated pieces of data without a data warehouse and without generating a snowflake data model, wherein the virtual schema is a multidimensional data model generated based on the data format of the one or more data sources;
an indexing engine and a scheduler that update the virtual schema when new pieces of data are received by the aggregator;
an analytical cube generator that generates a multidimensional analytical cube by executing commands from a declarative language against the virtual schema; and
an application programming interface to an external tool wherein the application programming interface permits the external tool to access the piece of data in the analytical cube.
2. The server of claim 1 further comprising one or more report definitions stored in the server, each report definition defining a report to be generated based on the pieces of data in the analytical cube.
3. The server of claim 1 further comprising a cache that stores the virtual schema.
4. A computer-implemented method for analyzing and reporting data from multiple data sources, the method comprising:
storing, in a server, one or more pieces of data from one or more data sources wherein the data sources include at least relational data sources and non-relational data sources;
converting, using one or more adapters that are associated with the server, one or more pieces of data from the data sources into a common data format;
generating, using an analytical cube generator associated with the server, a multidimensional analytical cube based on the one or more pieces of data in the common data format using a declarative language executed against a virtual schema that is a multidimensional data model generated based on the data format of the one or more data sources without a data warehouse and without generating a snowflake data model; and
providing, on the server, an application programming interface to an external tool wherein the application programming interface permits the external tool to access the piece of data in the analytical cube.
5. The method of claim 4 , wherein the external tool further comprises one or more of a business intelligence tool, a reporting application and an analytic tool.
6. The method of claim 4 further comprising receiving the one or more pieces of data from each data source using the adaptor that converts the one or more pieces of data from a particular data source having a particular format into the common data format.
7. The method of claim 4 , wherein the common data format further comprises XML.
8. The method of claim 7 , wherein the one or more data sources further comprises one or more of an XML data source, an HTML data source, an email data source, an excel data source, a PDF data source and a flat file data source.
9. The method of claim 4 , wherein the declarative language further comprises xPath.
10. A computer-implemented method for generating dimensional analytical models from extensible data, the method comprising:
receiving, at a server, one or more pieces of data from one or more data sources wherein the data sources include at least one of a relational data source and a non-relational data source;
converting, using one or more adapters that are associated with the server, the one or more pieces of data from the one or more data sources into a common data format;
forming, using an analytical cube generator associated with the server, a multidimensional analytical model for the one or more pieces of data stored in the common format; and
wherein forming the multidimensional analytical model further comprises using a declarative language to describe the multidimensional analytical model by executing commands from the declarative language against a virtual schema that is generated without a data warehouse and without generating a snowflake data model.
11. The method of claim 10 , wherein the declarative language further comprises xPath.
12. The method of claim 10 further comprising performing path analysis using the multidimensional analytical model wherein the path analysis is based on the declaration language.
13. The method of claim 12 , wherein performing path analysis further comprises uncovering data about a sequence of events contained in the one or more pieces of data.
14. The method of claim 10 further comprising identifying, using the multidimensional analytical model, co-occurrences between the one or more pieces of data.
15. The method of claim 10 further comprising determining an aggregated value for a particular piece of data from the one or more pieces of data wherein a first set of pieces of data has a first hierarchical structure that contains the particular piece of data and a second set of pieces of data has a second hierarchical structure that contains the particular piece of data at a different location than the first hierarchical structure, and wherein determining the aggregated value further comprises extracting the particular piece of data from each hierarchical structure using the multidimensional analytical model.