Method and system for handling multiple dimensions in relational databases

• US 20050076045A1
• Filed: 12/14/2001
• Published: 04/07/2005
• Est. Priority Date: 03/19/2001
• Status: Abandoned Application

• Current Assignee: Exie AS (AI Avocado Luxembourg Holding SARL)
• Original Assignee: Exie AS (AI Avocado Luxembourg Holding SARL)
• Inventors: Jensen, Brita Vefring, Lehne, Odd Arild, Stenslet, Pal
• Application Number:
  US10/471,802
• Publication Number:
  US 20050076045A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP

Abstract

The invention relates to a method for structuring data in a computerized relational database system. The method includes generating a dimension type table (dim_type) of descriptions of dimensions, a dimensional item table (dim_item) of dimensional items, a location table (location) of locations type representing intersection points between the various dimensions, and a connection table (dim_conn) defining connections between each location and dimensional items with which they are associated. Preferably, the values stored in the database are entered in a separate data value table (data_value). The invention also relates to a database system with data structured in accordance with the method. The invented data structure makes it possible to achieve flexibility and reuse of code without sacrificing scalability and performance of searching and reporting. Particularly, the structure facilitates a method for building queries for generating reports focusing on a subset of the data in the database limited by various criteria simultaneously involving an arbitrary number of dimensions, and a method for retrieving the items contained in a regular dimension classified according

Claims

1. Method for structuring data in a computerized relational database system comprising a storage device (5) for storing the relational database in the form of a number of tables, at least one central processor (1) connected to said storage device (5), and a database manager (3) comprising instructions capable of making the processor perform queries on said database, characterized by the steps of

generating a dimension type table (dim_type) of descriptions of the dimensions present in the database, each row of which including at least a dimension identifier (dim_type_id) and a dimension name (dim_type_name),

generating a dimensional item table (dim_item) of dimensional items, each row of which including at least an internal item identifier (dim_item_id), a presentation name (dim_item_name), and an identifier of a dimensional description (dim_type_id) in said dimension type table,

generating a location table (location) of locations representing intersection points between the various dimensions, each row of which including at least a location identifier (location_id), and

generating a connection table (dim_conn) defining connections between each location and dimensional items with which they are associated, each row of which including at least a location identifier (location_id) and a dimensional item identifier (dim_item_id)

2. Method according to claim 1, characterized in that it further comprises the step of generating a hierarchy table (dim_hier) defining dimensional hierarchies, each row of which including at least the identifier of an ancestor item (super_dim_item_id) in said dimensional item table (dim_item), the identifier of a descendant item (sub_dim_item_id) in said dimensional item table (dim_item), and the distance between the two items as number of hierarchical levels (distance), each item in said dimensional item table being defined as its own ancestor with distance zero.

3. Method according to claim 1 or 2, characterized in that each row of said location table (location) further includes one or more numeric or alphanumeric data values.

4. Method according to claim 1, characterized in that it further comprises the step of generating a data value table (data_value) of data contained in the database, each row of which including at least a location identifier (location_id) and a numeric or alphanumeric data value.

5. Method according to claim 1, characterized in that it further comprises the step of generating a dimensional item type table (dim_item_type) of descriptions of types that may be used in classification of items belonging to the same dimension, each row of which including at least an item type identifier (dim_item_type_id) and an item type name (dim_item_type_name);

and that each row of said dimensional item table (dim_item) further includes the identifier of an item type (dim_item_type_id) in said dimensional item type table (dim_item_type).

6. Method according to claim 1, characterized in that it further comprises the step of generating a connection type table (dim_conn_type) of definitions of different semantics that may be associated with a connection between a location and an item in a dimension, each row of which at least including a connection type identifier (dim_conn_type_id) and a connection type name (dim_conn_type_name);

and that each row of said connection table (dim_conn) further includes the identifier of a connection type (dim_conn_type_id) in said connection type table (dim_conn_type).

7. Method according to one of the claim 1, characterized in that it further comprises the step of generating a location type table (location_type) of classifications of locations in said location table (location), each row of which at least including a location type identifier (location_type_id) and a location type name (location_type_name);

and that each row of said location table (location) further includes the identifier of a location type (location_type_id) in said location type table (location_type).

33. Computer program product comprising instructions for, when loaded into and executed by a computer, causing the computer to perform the method of claim 1.

34. Computer program according to claim 33, embodied on a record medium.

35. Computer program according to claim 33, stored in a computer memory.

36. Computer program according to claim 33, carried on an electrical or optical carrier signal.

8. Computerized database system comprising a storage device (5) upon which is stored a relational database in the form of a number of tables, at least one central processor (1) connected to said storage device (5), and a database manager (3) comprising instructions capable of making the processor perform queries on said database, characterized by said tables including

a dimension type table (dim_type) of descriptions of the dimensions present in the database, each row of which including at least a dimension identifier (dim_type_id) and a dimension name (dim_type_name),

a dimensional item table (dim_item) of dimensional items, each row of which including at least an internal item identifier (dim_item_id), a presentation name (dim_item_name), and an identifier of a dimensional description (dim_type_id) in said dimension type table,

a location table (location) of locations representing intersection points between the various dimensions, each row of which including at least a location identifier (location_id), and

a connection table (dim_conn) defining connections between each location and dimensional items with which they are associated, each row of which including at least a location identifier (location_id) and a dimensional item identifier (dim_item_id).

9. System according to claim 8, characterized in that it further comprises a hierarchy table (dim_hier) defining dimensional hierarchies, each row of which including at least the identifier of an ancestor item (super_dim_item_id) in said dimensional item table (dim_item), the identifier of a descendant item (sub_dim_item_id) in said dimensional item table (dim_item), and the distance between the two items as number of hierarchical levels (distance), each item in said dimensional item table being defined as its own ancestor with distance zero.

10. System according to claim 8 or 9, characterized in that each row of said location table (location) further includes one or more numeric or alphanumeric data values.

11. System according to claim 8, characterized in that it further comprises a data value table (data_value) of data contained in the database, each row of which including at least a location identifier (location_id) and a numeric or alphanumeric data value.

12. System according to claim 8, characterized in that it further comprises a dimensional item type table (dim_item_type) of descriptions of types that may be used in classification of items belonging to the same dimension, each row of which including at least an item type identifier (dim_item_type_id) and an item type name (dim_item_type_name);

and that each row of said dimensional item table (dim_item) further includes the identifier of an item type (dim_item_type_id) in said dimensional item type table (dim_item_type).

13. System according to claim 8, characterized in that it further comprises a connection type table (dim_conn_type) of definitions of different semantics that may be associated with a connection between a location and an item in a dimension, each row of which at least including a connection type identifier (dim_conn_type_id) and a connection type name (dim_conn_type_name);

and that each row of said connection table (dim_conn) further includes the identifier of a connection type (dim_conn_type_id) in said connection type table (dim_conn_type).

14. System according to claim 8, characterized in that it further comprises a location type table (location_type) of classifications of locations in said location table (location), each row of which at least including a location type identifier (location_type_id) and a location type name (location_type_name);

and that each row of said location table (location) further includes the identifier of a location type (location_type_id) in said location type table (location_type).

15. Method for performing a computerized query in a relational database, where the database at least includes a dimension type table (dim_type) of dimensional types, a dimensional item table (dim_item) of dimensional items, a location table (location) of locations representing intersection points between the various dimensions, and a connection table (dim_conn) defining connections between each location and dimensional items with which they are associated, characterized by the steps of

specifying a query focus specification defining the subset of the database that the query should retrieve its results from consisting of a list of dimensional focus specifications, each specifying rules to identify a subset of dimensional items from one dimension of the database,

generating a first working table defining a query focus (query_focus), each row of which including a structural identifier (struct_id) identifying the dimensional focus specification the row belongs to, a focused item identifier (focus_dim_item_id) identifying an item in said dimensional item table (dim_item) of dimensional items that is in the query focus, and

performing a query in said relational database based on the focus defined in said first working table (query_focus), collecting data associated with locations connected to at least one of the focused items (focus_dim_item_id) identified for each structural identifier (struct_id) defined by the dimensional focus specifications contained in the query focus specification.

16. Method according to claim 15, characterized in that the step of specifying a query focus specification includes specifying for any given dimensional focus specification rules to directly or indirectly retrieve a list of explicitly selected item identifiers,

that the step of generating said first working table (query_focus) includes using an additional column for an explicitly selected item identifier (sel_dim_item_id) identifying an item in said dimensional item table (dim_item) of dimensional items that caused the item identified by the focused item identifier (focus_dim_item_id) to be considered part of the focus,

generating a second working table of query groups (query_group) defining groups that the data resulting from the query should be aggregated by each row of which including

a structural identifier (struct_id) and an explicitly selected item identifier (sel_dim_item_id) identifying rows in said first working table (query_focus) that should be included in the group,

a group identifier (group_id) that identifies the group, and

a group presentation name (group_pres_name) that gives the name to be used when presenting the group in the results of the aggregated query, and

that the step of performing a query in said relational database includes aggregating the results of the query according to the groups defined in said second working table (query_group).

17. Method according to claim 15 or 16, characterized in that the step of specifying a query focus specification includes specifying for any given dimensional focus specification, a list of explicitly selected item identifiers and a specification of generality of focus expressed as a number of ascendant and/or descendant hierarchical levels, and

that the step of generating said first working table for any given dimensional focus specification includes using a hierarchy table (dim_hier) defining dimensional hierarchies to determine the set of focused item identifiers (focus_dim_item_id) based on said list of explicitly selected item identifiers, and said specification of generality of focus.

18. Method according to claim 15, characterized in that the step of generating said first working table (query_focus) includes using a dimensional item type table (dim_item_type) describing types that may be used in classification of items belonging to the same dimension, to limit the set of focused item identifiers (focus_dim_item_id) identified for any given structural identifier (struct_id).

19. Method according to claim 15, characterized in that the step of performing said query includes using a connection type table (dim_conn_type) of definitions of different semantics that may be associated with connections between locations and items in a dimension, to collect data associated with locations connected with specific connection types (dim_conn_type_id) to at least one of the focused items (focus_dim_item_id) identified for any given structural identifier (struct_id).

20. Method according to claim 15, characterized in that the step of performing said query includes using a location type table (location_type) of classifications of locations in said location table (location), to collect data associated with locations with specific location types (dim_conn_type_id).

21. Method according to claim 15, characterized in that the step of generating said first working table (query_focus) includes iterating over the dimensional focus specifications contained in a given query focus specification and for each dimensional item determined to be in focus entering in the first working table (query_focus) a row containing,

the identification of the dimensional focus specification as a structural identifier (struct_id),

the dimensional item identifier (dim_item_id) of the item as said focused item identifier (focus_dim_item_id).

22. Method according to claim 16 characterized in that the step of generating said first working table (query_focus) includes iterating over the dimensional focus specifications contained in a given query focus specification and for each dimensional item determined to be in focus entering in the first working table (query_focus) a row containing,

the identification of the dimensional focus specification as a structural identifier (struct_id),

the dimensional item identifier (dim_item_id) of the item as said focused item identifier (focus_dim_item_id)

the dimensional item identifier (dim_item_id) of the dimensional item that caused said dimensional item to be in focus as said explicitly selected item identifier (sel_dim_item_id).

23. Method according to claim 22,

characterized in that each step of entering rows in said first working table is followed by a step determining whether the query result should be grouped by the explicitly selected items or some categorization of these, and if so, entering rows in the second working table (query_group) each row containing

the identification of the dimensional focus specification as said structural identifier (struct_id) and the dimensional item identifier (dim_item_id) of the explicitly selected item as said explicitly selected item identifier (sel_dim_item_id) as the two identifiers identifying rows in said first working table (query_focus),

a unique group identifier (group_id), and

a group presentation name (group_pres_name).

24. Method according to claim 23, characterized in that said group identifier (group_id) is the same as said explicitly selected item identifier (sel_dim_item_id).

25. Method according to claim 23, characterized in that said group identifier (group_id) is generated based on a categorization of the explicitly selected items.

26. Method according to claim 15, characterized in that said dimension type table (dim_type) contains one or more rows defining analysis dimensions,

that said dimensional item table (dim_item) contains one or more rows defining analysis items belonging to said analysis dimensions,

that the database contains definitions of connections between items in said analysis dimensions and items belonging to regular dimensions, and

that the step of specifying a query focus specification for any given dimensional focus specification on a regular dimension includes specifying an analysis focus specification consisting of a list of dimensional focus specifications on analysis dimensions, each specifying rules to identify a subset of analysis items from an analysis dimension.

27. Method according to claim 22, characterized in that said dimension type table (dim_type) contains one or more rows defining analysis dimensions,

that said dimensional item table (dim_item) contains one or more rows defining analysis items belonging to said analysis dimensions,

that the database contains definitions of connections between items in said analysis dimensions and items belonging to regular dimensions, and

that said dimensional focus specification contains an analysis focus specification consisting of a list of dimensional focus specifications on analysis dimensions, each specifying rules to identify a subset of analysis items from an analysis dimension, and

that the list of explicitly selected items for a dimensional focus specification is retrieved from the database by applying said analysis focus specification.

28. Method according to claim 23, characterized in that said dimension type table (dim_type) contains one or more rows defining analysis dimensions,

that said dimensional item table (dim_item) contains one or more rows defining analysis items belonging to said analysis dimensions,

that the database contains definitions of connections between items in said analysis dimensions and items belonging to regular dimensions, and

that said dimensional focus specification contains an analysis focus specification consisting of a list of dimensional focus specifications on analysis dimensions, each specifying rules to identify a subset of analysis items from an analysis dimension, and

that the step of entering rows in second working table (query_group) generates group identifiers (group_id) and group presentation names (group_pres_name) by retrieving from the database information about analysis items connected to the explicitly selected items identified by applying said analysis focus specification.

37. Computer program product comprising instructions for, when loaded into and executed by a computer, causing the computer to perform the method of claim 15.

38. Computer program according to claim 37, embodied on a record medium.

39. Computer program according to claim 37, stored in a computer memory.

40. Computer program according to claim 37, carried on an electrical or optical carrier signal.

41. Computer program product comprising instructions for, when loaded into and executed by a computer, causing the computer to perform the method of claim 28.

42. Computer program according to claim 41, embodied on a record medium.

43. Computer program according to claim 41, stored in a computer memory.

44. Computer program according to claim 41, carried on an electrical or optical carrier signal.

29. Method for performing a computerized query in a relational database systems, where the database at least includes a dimension type table (dim_type) of dimensional types, a dimensional item table (dim_item) of dimensional items, a location table (location) of locations representing intersection points between the various dimensions, and a connection table (dim_conn) defining connections between each location and dimensional items with which they are associated, characterized by the steps of

entering in said dimension type table (dim_type) one or more rows defining analysis dimensions,

entering in said dimensional item table (dim_item) one or more rows defining analysis items belonging to said analysis dimensions,

defining connections between items in said analysis dimensions and items belonging to regular dimensions, and

performing queries that limit and/or aggregate results according to specifications on the analysis dimensions.

30. Method according to claim 29, characterized in using a hierarchy table (dim_hier) defining dimensional hierarchies within analysis dimensions that are hierarchical.

31. Method according to claim 29 or 30, characterized in that said connections between items in said analysis dimensions and items in other dimensions are defined by

entering in a location type table (location_type) containing classifications of locations in said location table (location), a row that defines identity,

entering in said location table (location) rows representing one-to-one each item in any regular dimension that are to be accessed through the analysis dimension, each such row also referring to the entry in said location type table (location_type) defining identity,

entering in said connection table (dim_conn) one row for each connection between a location in said location table (location) and the regular dimensional item it represents in said dimensional item table (dim_item),

entering in said connection table (dim_conn) one row for each connection between a location in said location table (location) representing an item in a regular dimension and an item belonging to an analysis dimension in said dimensional item table (dim_item).

32. Method according to claim 31,

of using a first regular dimension as an analysis dimension towards a second regular dimension characterized in

generating a connection type table (dim_conn_type) of definitions of different semantics that may be associated with a connection between a location and an item in a dimension;

and that each row of said connection table (dim_conn) further includes the identifier of a connection type in said connection type table (dim_conn_type),

entering in said connection type table (dim_conn_type) one or more rows representing connection types indicating the use of said first regular dimension as an analysis dimension towards said second regular dimension,

entering in said connection table (dim_conn) one row for each connection between a location in said location table (location) representing an item in said second regular dimension and an item belonging to said first regular dimension dimension in said dimensional item table (dim_item).