Sheaf data model

• US 20050192986A1
• Filed: 04/22/2005
• Published: 09/01/2005
• Est. Priority Date: 05/12/2000
• Status: Active Grant

• Current Assignee: Limit Point Systems Incorporated
• Original Assignee: Limit Point Systems Incorporated
• Inventors: Butler, David M.
• Granted Patent: US 7,865,526 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/288 Entity relationship models
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure

Abstract

A data model including a first table to represent an entity type with a column in the table for a respective attribute of the entity type, and having rows entered with attribute data. The data model also includes a row graph which represents an ordering relationship between the rows of the first table, and a column graph which represents an ordering relationship between columns of the first table. Further, the column graph is a row graph from a second table.

Claims

1. A computer-implemented method of representing data on a computer as a collection of finite sheaves for facilitating information retrieval in response to a query, comprising the steps of:

selecting one or more entity sets within an application, each entity set being a collection of entities of a single entity type;

constructing a distinct table to represent each entity set selected in said selecting step, such that there is a one-to-one correspondence between respective tables and entity sets;

for each table executing steps of

selecting another table and a member in a finite distributive lattice, FDL, defined by said table, hereafter referred to as a schema table and a schema member, respectively, of a current table,

constructing a column in the current table for each row in a down set of the respective schema member, such that there is a one-to-one correspondence between columns in the current table and rows in the down set of the schema member,

interpreting each column of the table as an attribute of the associated entity type,

interpreting a row graph of the down set of the schema member as a column graph of the current table, said column graph representing an ordering relationship between the columns of the current table,

constructing a row in the current table for each entity of the associated entity set, such that there is a one-to-one correspondence between rows and entities,

entering attribute data into rows of the current table,

imposing an externally specified ordering relationship between the rows of the current table, hereafter referred to as the row graph,

defining the FDL from the row graph to have a member for each row and each distinct combination of the rows of the current table, wherein two combinations of rows are distinct if they do not represent a same ordering relationship, and

defining a finite sheaf from the table, row graph and column graph.

2. The method according to claim 1, wherein a first entity is included in a second entity if and only if there is a path in the graph from a node corresponding to the first entity to a node corresponding to the second entity.

3. The method according to claim 1, wherein the ordering relationship of the rows of the first table comprises a partially ordered relationship.

4. The method according to claim 1, further comprising steps of executing commands indicative of operations for querying and manipulating the table, row graph, and column graph of one finite sheaf or the respective tables, row graphs, and column graphs of several finite sheaves in combination.

5. The method according to claim 1, wherein the attribute data comprises at least one of simulation data, spatial data, object-orientated data, and relational data.

6. A data model encoded on a computer-readable medium as a collection of finite sheaves for facilitating information retrieval in response to a query, comprising:

a first table and other tables respectively configured to represent a single entity set from a group of entity sets, such that there is a one-to-one correspondence between respective tables and entity sets, said first table having

an association with a second table and a member in a finite distributive lattice, FDL, defined by said table, hereafter referred to as the schema table and schema member, respectively, of the first table,

a column for each row in a down set of the respective schema member, such that there is a one-to-one correspondence between columns in the first table and rows in the down set of the schema member, each said column corresponding to an attribute of the entity type associated with the first table,

a column graph defined to be a row graph of the down set of the respective schema member, said column graph representing an ordering relationship between the columns of the current table,

a row for each entity of the associated entity set, such that there is a one-to-one correspondence between rows and entities,

attribute data included in the rows,

an externally specified ordering relationship between the rows of the first table, hereafter referred to as the row graph,

the FDL defined from the row graph to have a member for each row and each distinct combination of the rows of the first table, wherein two combinations of rows are distinct if they do not represent a same ordering relationship, and

a finite sheaf corresponding to the table, row graph and column graph.

7. The data model according to claim 6, wherein a first entity is included in a second entity if and only if there is a path in the graph from a node corresponding to the first entity to a node corresponding to the second entity.

8. The data model according to claim 6, wherein the ordering relationship of the rows of the first table comprises a partially ordered relationship.

9. The data model according to claim 6, further comprising commands indicative of operations for querying and manipulating the table, row graph, and column graph of one finite sheaf or the respective tables, row graphs, and column graphs of several finite sheaves in combination.

10. The data model according to claim 6, wherein the attribute data comprises at least one of simulation data, spatial data, object-orientated data, and relational data.