Performing cross-tabulation using a columnar database management system

1. A columnar database data mining method, comprising:

receiving, by a columnar database management system from a requestor, a request to perform a cross-tabulation operation on input tables having a columnar format, the request specifying row dimensions representing columns of the input tables, the columnar database management system having a processor and a non-transitory computer-readable medium;

responsive to the request, determining, by the columnar database management system, a result table comprising rows, each row of the result table having a combination of values of the row dimensions, the determining comprising, for each row of the result table:

for each value of a row dimension, determining a domain comprising a set of records from an input table of the input tables, the set of records matching the each value of the row dimension;

determining an intersection set of domains corresponding to the each row, the intersection set comprising a set of records that match the combination of values of the row dimensions; and

determining a value for a result column for the each row as an aggregate value based on the intersection set; and

storing the result table in a data store or returning a response to the requestor.

2. The columnar database data mining method according to claim 1 , wherein the request specifies a first row dimension representing a column of a first table of the input tables and wherein determining a domain for each value of a row dimension comprises selecting the set of records from the first table that meet a given condition or value of the column of the first table.

3. The columnar database data mining method according to claim 2 , wherein the request specifies a second row dimension representing a column of a second table, the second table having a column keyed to the first table.

4. The columnar database data mining method according to claim 1 , wherein the request is received from a client device through an application frontend, directly by the database engine of the columnar database management system, via an application programming interface call, through receipt of a file containing the request, or via an interactive console.

5. The columnar database data mining method according to claim 1 , wherein the result column comprises a pivot dimension, the method further comprising, for each value of the pivot dimension:

determining a pivot domain having a set of records that meet the each value of the pivot dimension; and

determining an intersection of the pivot domain with domains corresponding to the values of the row dimensions.

6. The columnar database data mining method according to claim 1 , wherein the result column comprises comparison values determined by applying an operator to results of a base filter and results of a target filter and wherein the operator is one of a difference operator, a relative index operator, or a percent operator.

7. The columnar database data mining method according to claim 1 , wherein the aggregate value represents a measure produced by an aggregation function utilizing values corresponding to the intersection set.

8. A system, comprising:

a processor;

a non-transitory computer-readable medium; and

stored instructions translatable by the processor to perform:

receiving, from a requestor, a request to perform a cross-tabulation operation on input tables having a columnar format, the request specifying row dimensions representing columns of the input tables;

responsive to the request, determining a result table comprising rows, each row of the result table having a combination of values of the row dimensions, the determining comprising, for each row of the result table:

for each value of a row dimension, determining a domain comprising a set of records from an input table of the input tables, the set of records matching the each value of the row dimension;

determining an intersection set of domains corresponding to the each row, the intersection set comprising a set of records that match the combination of values of the row dimensions; and

determining a value for a result column for the each row as an aggregate value based on the intersection set; and

storing the result table in a data store or returning a response to the requestor.

9. The system of claim 8 , wherein the request specifies a first row dimension representing a column of a first table of the input tables and wherein determining a domain for each value of a row dimension comprises selecting the set of records from the first table that meet a given condition or value of the column of the first table.

10. The system of claim 9 , wherein the request specifies a second row dimension representing a column of a second table, the second table having a column keyed to the first table.

11. The system of claim 8 , wherein the request is received from a client device through an application frontend, directly by the database engine of the columnar database management system, via an application programming interface call, through receipt of a file containing the request, or via an interactive console.

12. The system of claim 8 , wherein the result column comprises a pivot dimension, the method further comprising, for each value of the pivot dimension:

determining a pivot domain having a set of records that meet the each value of the pivot dimension; and

determining an intersection of the pivot domain with domains corresponding to the values of the row dimensions.

13. The system of claim 8 , wherein the result column comprises comparison values determined by applying an operator to results of a base filter and results of a target filter and wherein the operator is one of a difference operator, a relative index operator, or a percent operator.

14. The system of claim 8 , wherein the aggregate value represents a measure produced by an aggregation function utilizing values corresponding to the intersection set.

15. A computer program product for columnar database data mining, the computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor to perform:

receiving, from a requestor, a request to perform a cross-tabulation operation on input tables having a columnar format, the request specifying row dimensions representing columns of the input tables;

responsive to the request, determining a result table comprising rows, each row of the result table having a combination of values of the row dimensions, the determining comprising, for each row of the result table:

for each value of a row dimension, determining a domain comprising a set of records from an input table of the input tables, the set of records matching the each value of the row dimension;

determining an intersection set of domains corresponding to the each row, the intersection set comprising a set of records that match the combination of values of the row dimensions; and

determining a value for a result column for the each row as an aggregate value based on the intersection set; and

storing the result table in a data store or returning a response to the requestor.

16. The computer program product of claim 15 , wherein the request specifies a first row dimension representing a column of a first table of the input tables and wherein determining a domain for each value of a row dimension comprises selecting the set of records from the first table that meet a given condition or value of the column of the first table.

17. The computer program product of claim 16 , wherein the request specifies a second row dimension representing a column of a second table, the second table having a column keyed to the first table.

18. The computer program product of claim 15 , wherein the request is received from a client device through an application frontend, directly by the database engine of the columnar database management system, via an application programming interface call, through receipt of a file containing the request, or via an interactive console.

19. The computer program product of claim 15 , wherein the result column comprises a pivot dimension, the method further comprising, for each value of the pivot dimension:

determining a pivot domain having a set of records that meet the each value of the pivot dimension; and

determining an intersection of the pivot domain with domains corresponding to the values of the row dimensions.

20. The computer program product of claim 15 , wherein the result column comprises comparison values determined by applying an operator to results of a base filter and results of a target filter and wherein the operator is one of a difference operator, a relative index operator, or a percent operator.