Many-core algorithms for in-memory column store databases

1. A non-transitory computer program product storing instructions that, when executed by at least one programmable processor, cause the at least one programmable processor to perform operations comprising:

receiving a query request comprising a definition of a query of a database persisted in a column-based storage;

identifying a pattern in at least part of the query;

matching the identified pattern with a set of pre-defined patterns, each of the pre-defined patterns having associated therewith at least one pre-compiled query execution sub-component of a plurality of pre-compiled query execution sub-components retained in a library;

selecting, based on the matching of the identified patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query that avoids intermediate results and allows for the query to be performed in a single pass of the database, wherein at least one matching pattern other than the selected matched pattern produces intermediate results that requires more than two passes of the database;

generating, based on the matching, a plan for executing the query; and

executing the query using the generated plan.

2. A computer program product as in claim 1 , wherein the operations further comprise

selecting, based on the matching of the identified patterns with the set of pre-defined patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query; and

configuring the plan to include the optimal sequence for processing the plurality of tables.

3. A computer program product as in claim 1 , wherein the generating comprises incorporating the pre-compiled query execution sub-component associated with the matched pattern into the plan based on a pseudo code representation of the plan derived from the definition.

4. A computer program product as in claim 3 , wherein the operations further comprise deriving the pseudo code representation of the plan from the definition.

5. A computer program product as in claim 1 , wherein the generating comprises creating a single function to call the pre-compiled query execution component and the one or more other pre-compiled query execution components to generate the plan.

6. A computer program product as in claim 5 , wherein the single function defines a desired result and accesses a predefined parallelization plan from a set of two or more predefined parallelization plans based at least in part of the matching of the identified pattern.

7. A computer program product as in claim 1 , wherein the pre-compiled query execution sub-component comprises one or more pre-compiled SQL operations expressed in C++.

8. A system comprising:

at least one programmable processor; and

a machine-readable medium storing instructions that, when executed by the at least one processor, cause the at least one programmable processor to perform operations comprising:

receiving a query request comprising a definition of a query of a database persisted in a column-based storage;

identifying a pattern in at least part of the query;

matching the identified pattern with a set of pre-defined patterns, each of the pre-defined patterns having associated therewith at least one pre-compiled query execution sub-component of a plurality of pre-compiled query execution sub-components retained in a library;

selecting, based on the matching of the identified patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query that avoids intermediate results and allows for the query to be performed in a single pass of the database, wherein at least one matching pattern other than the selected matched pattern produces intermediate results that requires more than two passes of the database;

generating, based on the matching, a plan for executing the query; and

executing the query using the generated plan.

9. A system as in claim 8 , wherein the operations further comprise:

selecting, based on the matching of the identified patterns with the set of pre-defined patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query; and

configuring the plan to include the optimal sequence for processing the plurality of tables.

10. A system as in claim 8 , wherein the generating comprises incorporating the pre-compiled query execution sub-component associated with the matched pattern into the plan based on a pseudo code representation of the plan derived from the definition.

11. A system as in claim 10 , wherein the operations further comprise deriving the pseudo code representation of the plan from the definition.

12. A system as in claim 8 , wherein the generating further comprises creating a single function to call the pre-compiled query execution component and the one or more other pre-compiled query execution components to generate the plan.

13. A system as in claim 12 , wherein the single function defines a desired result and accesses a predefined parallelization plan from a set of two or more predefined parallelization plans based at least in part of the matching of the identified pattern.

14. A system as in claim 8 , wherein the pre-compiled query execution sub-component comprises one or more pre-compiled SQL operations expressed in C++.

15. A computer-implemented method comprising:

receiving a query request comprising a definition of a query of a database persisted in a column-based storage;

identifying a pattern in at least part of the query;

matching the identified pattern with a set of pre-defined patterns, each of the pre-defined patterns having associated therewith at least one pre-compiled query execution sub-component of a plurality of pre-compiled query execution sub-components retained in a library;

selecting, based on the matching of the identified patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query that avoids intermediate results and allows for the query to be performed in a single pass of the database, wherein at least one matching pattern other than the selected matched pattern produces intermediate results that requires more than two passes of the database;

generating, based on the matching, a plan for executing the query; and

executing the query using the generated plan.

16. A computer-implemented method as in claim 15 , further comprising:

selecting, based on the matching of the identified patterns with the set of pre-defined patterns, an optimal sequence for processing a plurality of tables that must be joined to respond to the query; and

configuring the plan to include the optimal sequence for processing the plurality of tables.

17. A computer-implemented method as in claim 15 , further comprising deriving a pseudo code representation of the plan from the definition.

18. A computer-implemented method as in claim 15 , wherein the generating further comprises creating a single function to call the pre-compiled query execution component and the one or more other pre-compiled query execution components to generate the plan.

19. A computer-implemented method as in claim 18 , wherein the single function defines a desired result and accesses a predefined parallelization plan from a set of two or more predefined parallelization plans based at least in part of the matching of the identified pattern.

20. A computer-implemented method as in claim 15 , wherein the pre-compiled query execution sub-component comprises one or more pre-compiled SQL operations expressed in C++, wherein at least one of the receiving, the identifying, the matching, the generating, and the executing is performed by at least one programmable processor.