Multidimensional domain modeling method and system

• US 5,918,232 A
• Filed: 11/26/1997
• Issued: 06/29/1999
• Est. Priority Date: 11/26/1997
• Status: Expired due to Term

• Current Assignee: SymphonyRPM, Inc. (Samsung Electronics Co. Ltd.)
• Original Assignee: WhiteLight Systems, Inc. (Samsung Electronics Co. Ltd.)
• Inventors: Brill, Michael L., Stross, Kenner G., Pouschine, Nicholas
• Primary Examiner(s): Amsbury, Wayne
• Assistant Examiner(s): LEWIS, CHERYL RENEA
• Application Number:
  US08/978,168
• Time in Patent Office: 580 Days
• Field of Search: 707/103, 707/2, 707/3, 707/4
• US Class Current: 1/1
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99934 Query formulation, input preparation, or translation
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure

Abstract

A system and method for computer modeling (10) and for creating hyperstructures (51) which are to be contained in a computer memory, which obtains measurements of physical objects and activities which are related to the entity to be modeled in the computer hyperstructure (51). The measurements are transformed into computer data which corresponds to the physical objects and activities external to the computer system (10). A plurality of independent dimensions (54) are created, where each dimension (54) includes at least one element (58). A plurality of cells (56) are created, each of which is associated with the intersection of two or more elements (58), each cell (56) being capable of storing at least one value. At least one rule domain (60) is associated with at least one cell (56), the rule domain (60) including at least one rule for assigning values to the associated cells (56). A domain modeling rule set (126) is prepared (300), which determines which of the rules will provide the value associated with each of the cells (56) wherein application of the domain modeling rule set (126) to the hyperstructure (51) causes a physical transformation of the data corresponding to said physical objects which are modeled in said hyperstructure (51).

Also disclosed is a method for querying computer hyperstructures (51), a Hyperstructure Query Language, and a "cell explorer", which allows direct viewing of the applied formulas that produce a specific value for a cell (56).

Claims

1. A method of modeling a plurality of variables in a hyperstructure in a computer modeling system having at least one data storage means, a data processing means, input means, and output means, the steps comprising:

(A) obtaining measurements of physical objects and activities which are related to the entity to be modeled in the computer hyperstructure;

(B) transforming said measurements into computer data which corresponds to the physical objects and activities external to the computer system;

(C) constructing a plurality of independent dimensions from said computer data within a hyperstructure, where each dimension has at least one element;

(D) creating a plurality of cells, each of which is associated with the intersection of at least two elements, each cell being capable of storing at least one value;

(E) associating at least one rule domain with at least one cell, said rule domain including at least one means for assigning values to the associated cells; and

(F) preparing a domain modeling rule set which determines which of said assigning means will provide the value associated with each of said cells, wherein the application of said domain modeling rule set to the hyperstructure causes a physical transformation of the data corresponding to said physical objects which are modeled in said hyperstructure.

2. The method of computer modeling of claim 1 wherein:

the means of assigning values in step (E) is a rule selected from the group consisting of User Entered Formulas, autoformulas, User Entered Values, and datamaps.

3. The method of computer modeling of claim 1, wherein step (F) further comprises:

creating a Do Not Aggregate List, a Datamap Rules List, an Autoformula List and a Rules List.

4. The method of computer modeling of claim 3 wherein step (F) further comprises:

(a) gathering the rules;

(b) processing the autoformulas;

(c) processing the datamaps;

(d) ordering the rules;

(e) promoting the rules; and

(f) processing rule conflicts.

5. The computer modeling method of claim 4, wherein step (a) includes:

(1) building a Do Not Aggregate List;

(2) getting the next rule;

(3) if the rule is a datamap, then saving the datamap in the Datamap Rules List;

(4) if the rule is an autoformula, then adding the rule to the Autoformula List;

(5) if the rule is an aggregation, then creating a time summary property rule and adding the time summary property rule to the Rules List;

(6) else adding the rule to the Rules List; and

(7) repeating steps (2) through (6) until the last rule has been processed.

6. The computer modeling method of claim 5, wherein step (5) includes:

(a) getting the next time summary property (TSP) from the model;

(b) creating a new rule for that TSP;

(c) adding this new rule to the Rules List; and

(d) repeating steps (a) through (c) until all TSPs have been processed, and returning.

7. The computer modeling method of claim 6, wherein step (c) includes:

(i) if the rule is not a list formula, then appending the rule to the Rules List, and returning;

(ii) if the rule is a list formula, then fetching the first next element in the rule domain associated therewith;

(iii) if said first next element does not have a "Do not aggregate" property, then fetching the second next element;

(iv) if said first next element has a "Do not aggregate" property, and said first next element is in this rule'"'"'s dimension, then fetching the second next element, else removing said first next element from the formula'"'"'s domain; and

(v) repeating steps (ii) through (iv) until all elements have been processed, then appending the rule to the Rules List and returning.

8. The computer modeling method of claim 4, wherein step (b) includes:

(1) getting the next rule in the Autoformula List;

(2) if the rule is not in the Time dimension, then adding the rule to the Rules List;

(3) else creating a time summary property rule and adding the time summary property rule to the Rules List; and

(4) repeating steps (1)-(3) until the last rule has been processed.

9. The computer modeling method of claim 4, wherein step (c) includes:

(1) gathering hierarchies and levels from all dimensions;

(2) getting the next rule from the Datamap Rules List;

(3) setting all elements and dimensions as valid;

(4) getting the next hierarchy;

(5) getting the next level;

(6) if this level is joinable to the previous level, then

(a) if the level is first in its dimension, then turning off all elements in this dimension and adding this level'"'"'s elements to the rule'"'"'s domain;

(b) if the level is not first in its dimension, then adding the level'"'"'s elements to the rule'"'"'s domain;

(c) in either case (a) or (b), then

(i) if datamap is a low level datamap, then getting the next hierarchy;

(ii) if datamap is not a low level datamap, then getting the next level;

(7) if this level is not joinable to the previous level, then getting the next level;

(8) repeating steps (5), (6) and (7) until all levels in a hierarchy are processed;

(9) repeating steps (4)-(8) until all hierarchies in a rule are processed, then turning off all elements not in the table'"'"'s rule domain and adding the rule to the Rules List; and

(10) repeating steps (2)-(9) until all rules are processed.

10. The computer modeling method of claim 4, wherein step (d) includes:

(1) getting the next rule from the Rules List;

(2) assigning a primary sort value depending on rule type;

(3) assigning a secondary sort value depending on rule type;

(4) repeating steps (1)-(3) until each rule has been assigned primary and secondary sort values; and

(5) ordering the Rules List by primary and secondary sort values.

11. The computer modeling method of claim 4, wherein step (e) includes:

(1) establishing a Trigger List;

(2) getting the next rule from the Rules List;

(3) if the rule is a user entered value, or a user entered formula, then adding the rule to the Trigger List;

(4) else getting the next rule from the Rules List;

(5) repeating steps (2)-(4) until the last rule has been processed;

(6) getting the next rule from the Trigger List;

(7) getting the next autoformula from the Autoformula List;

(8) computing the affected elements from the autoformula'"'"'s dimension which are superceded by the Trigger List to make a modified autoformula list;

(9) repeating steps (6) and (7) until all autoformulas are processed;

(10) getting the next autoformula from the modified autoformula list;

(11) creating a promoted autoformula rule;

(12) repeating steps (10) and (11) until all autoformulas are processed; and

(13) repeating steps (6)-(12) until all rules have been processed.

12. The computer modeling method of claim 4, wherein step (f) includes:

(1) getting the next rule from the Rules List;

(2) creating a check set consisting of all user entered values with equal specificity;

(3) creating a check set consisting of all high priority rules with equivalent specificity;

(4) creating a check set consisting of all non-commutative rules with low priority with equivalent specificity;

(5) creating a check set consisting of all remaining commutative, low priority rules with equivalent specificity from the same dimension;

(6) if any rule'"'"'s domain intersects any rules in the check sets, then inserting a conflict rule with a rule domain equal to the intersection region ahead of the rule in the list, and saving references to the conflict rule;

(7) if there is no intersection, then getting the next rule from the Rules List; and

(8) repeating steps (1)-(7) until all rules have been processed.

13. The method of computer modeling of claim 1, the steps further comprising:

(G) providing a computer data manipulation language suitable for expressing queries to obtain data from a hyperstructure.

14. The method of computer modeling of claim 1, the steps further:

(H) providing a user interface means by which the method of calculation of each cell value can be observed.

15. A method of querying a multidimensional computer modeling data structure in a computer system having at least one data storage means, data processing means, input means, output means, a hyperstructure constructed in one of said data storage means, a calculation engine, a domain modeling rule set preparation module, a query engine, and an evaluator which communicates with an SQL generator, a math library, a cache retrieval device and a sort and search processor, comprising the steps of:

(A) inputting a query;

(B) parsing the query;

(C) creating a query component tree;

(D) inputting model metadata into said domain modeling rule set preparation module;

(E) generating a domain modeling rule set from said domain modeling rule set preparation module;

(F) inputting said domain modeling rule set into said calculation engine;

(G) generating an execution tree with rules from said calculation engine;

(H) inputting said execution tree with rules to said query engine;

(I) generating an optimized execution tree from said query engine;

(J) inputting said optimized execution tree to said evaluator;

(K) communicating between said evaluator and a data reference means;

(L) generating an execution tree with results;

(M) repeating steps (C)-(L) if there had been any inner queries presented before the query as a whole can be addressed;

(N) packaging the query results; and

(O) outputting query results.

16. The method of querying of claim 15, wherein:

said data reference means is selected from the group consisting of an SQL generator which communicates with a Relational Database Management System, a math library, a sort and search processor, and a cache retrieval means which communicates with a multi-dimension cache.

17. The method of querying of claim 15, wherein step (E) further comprises the steps of:

(a) gathering the rules;

(b) processing the autoformulas;

(c) processing the datamaps;

(d) ordering the rules;

(e) promoting the rules; and

(f) processing rule conflicts.

18. A computer system for modeling a plurality of variables in a hyperstructure comprising:

at least one data storage means;

data processing means;

input means;

output means;

measurements of physical objects and activities which are related to the entity to be modeled in the computer hyperstructure which are transformed into computer data which corresponds to the physical objects and activities external to the computer system;

a plurality of independent dimensions within the hyperstructure constructed from said data, where each dimension has at least one element;

a plurality of cells, each of which is associated with the intersection of at least two elements, each cell being capable of storing at least one value;

at least one rule domain associated with at least one cell, each said rule domain including at least one means for assigning values to the associated cell; and

a domain modeling rule set for determining which of said assigning means will provide the value associated with each of said cells, wherein the application of said domain modeling rule set to the hyperstructure causes a physical transformation of the data corresponding to said physical objects which are modeled in said hyperstructure.