STATIC DATA CACHING FOR QUERIES WITH A CLAUSE THAT REQUIRES MULTIPLE ITERATIONS TO EXECUTE

• US 20170116271A1
• Filed: 09/16/2016
• Published: 04/27/2017
• Est. Priority Date: 10/23/2015
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Ziauddin, Mohamed, Zhu, Yali
• Granted Patent: US 10,642,831 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 16/24539 using cached or materialised query results
• CPC Class Codes: G06F 16/24542 Plan optimisation
• CPC Class Codes: G06F 16/24566 Recursive queries

Abstract

Techniques are described herein to generate and to execute a query execution plan using static data buffering. After receiving a query with a clause that requires multiple iterations to execute, a database management system (DBMS) generates a plurality of plans that vary the order in which the database operations are executed. Within each plan, the DBMS identifies sets of rows within that plan that contain static data during execution of the query. Then, an additional step is added to each plan that includes loading the static set of rows in a database buffer cache. One or more database operations, from an iteration other than the first iteration, may be performed against the cached static set of rows. For each plan generated in this manner, a cost analysis model is applied, and the plan with the lowest estimated computational cost is selected for use as the query execution plan.

Claims

1. A method comprising:

receiving a query with a clause that requires multiple iterations to execute;

prior to executing the query, determining, based on the query, one or more database operations that are;

performed in more than one iteration, and

at least partially performed against a static set of rows, said static set of rows being static during execution of the query;

while executing the query,

loading said static set of rows into a database buffer cache; and

performing at least one iteration other than a first iteration against the buffered static set of rows.

2. The method of claim 1, further comprising:

projecting one or more columns from a particular static set of rows from a database object that is used as input for a database operation in the query; and

after performing the column projection, using a resulting static set of rows as the static set of rows to load into the database buffer cache.

3. The method of claim 1, further comprising:

applying one or more filter predicates to a particular static set of rows from a database object that is used as input for a database operation in the query; and

after applying the one or more filter predicates, using a resulting static set of rows as the static set of rows to load into the database buffer cache.

4. The method of claim 1, further comprising:

applying a join operation to a first static set of rows from a first database object and a second static set of rows from a second database object; and

after applying the join operation, using a resulting static set of rows as the static set of rows to load into the database buffer cache.

5. The method of claim 1, further comprising:

prior to the executing the clause that requires multiple iterations to execute,

generating a plan to execute the clause that includes;

performing a first database operation that produces a first static set of rows; and

after performing the first database operation, using the first static set of rows as input to perform a second database operation that produces a second static set of rows;

estimating a first computational cost of executing the plan by buffering the first static set of rows;

estimating a second computational cost of executing the plan by buffering the second static set of rows; and

generating a query execution plan based on the lowest estimated computational cost.

6. The method of claim 5, further comprising:

generating a plurality of query execution plans that vary an order in which the database operations are executed.

7. The method of claim 1, wherein executing the query is performed by a plurality of database server instances;

wherein each database server instance has a respective database buffer cache of a plurality of database buffer caches;

wherein the method further comprises distributing the static set of rows across the plurality of database buffer caches.

8. The method of claim 7, wherein each database server instance includes at least one respective slave process to execute the query;

wherein the method further comprises locking the respective slave processes from executing other queries while executing the query.

9. One or more non-transitory computer-readable media storing one or more sequences of instructions which, when executed by one or more processors, cause:

receiving a query with a clause that requires multiple iterations to execute;

prior to executing the query, determining, based on the query, one or more database operations that are;

performed in more than one iteration, and

at least partially performed against a static set of rows, said static set of rows being static during execution of the query;

while executing the query,

loading said static set of rows into a database buffer cache; and

performing at least one iteration other than a first iteration against the buffered static set of rows.

10. The one or more non-transitory computer-readable media of claim 9, storing instructions which, when executed by the one or more processors, further cause:

applying one or more filter predicates to a particular static set of rows from a database object that is used as input for a database operation in the query; and

after applying the one or more filter predicates, using a resulting static set of rows as the static set of rows to load into the database buffer cache.

11. The one or more non-transitory computer-readable media of claim 9, storing instructions which, when executed by the one or more processors, further cause:

applying a join operation to a first static set of rows from a first database object and a second static set of rows from a second database object; and

after applying the join operation, using a resulting static set of rows as the static set of rows to load into the database buffer cache.

12. The one or more non-transitory computer-readable media of claim 9, storing instructions which, when executed by the one or more processors, further cause:

prior to the executing the clause that requires multiple iterations to execute,

generating a plan to execute the clause that includes;

performing a first database operation that produces a first static set of rows; and

after performing the first database operation, consuming the first static set of rows to perform a second database operation that produces a second static set of rows;

estimating a first computational cost of executing the plan by buffering the first static set of rows;

estimating a second computational cost of executing the plan by buffering the second static set of rows; and

generating a query execution plan based on the lowest estimated computational cost.

13. The one or more non-transitory computer-readable media of claim 9, storing instructions which, when executed by the one or more processors, further cause:

generating a plurality of query execution plans that vary an order in which the database operations are executed.

14. The one or more non-transitory computer-readable media of claim 9, wherein executing the query is performed by a plurality of database server instances;

wherein each database server instance has a respective database buffer cache of a plurality of database buffer caches;

wherein the one or more non-transitory computer-readable media stores instructions which, when executed by the one or more processors, further cause;

distributing the static set of rows across the plurality of database buffer caches.

15. The one or more non-transitory computer-readable media of claim 14, wherein each database server instance includes at least one respective slave process to execute the query;

wherein the one or more non-transitory computer-readable media stores instructions which, when executed by the one or more processors, further cause;

locking the respective slave processes from executing other queries while executing the query.

16. A database management system comprising one or more computing devices configured to:

receive a query with a clause that requires multiple iterations to execute;

prior to executing the query, determine, based on the query, one or more database operations that are;

performed in more than one iteration, and

at least partially performed against a static set of rows, said static set of rows being static during execution of the query;

while executing the query,

load said static set of rows into a database buffer cache; and

perform at least one iteration other than a first iteration against the buffered static set of rows.

17. The database management system of claim 16, comprising the one or more computing devices further configured to:

apply a join operation to a first static set of rows from a first database object and a second static set of rows from a second database object; and

after applying the join operation, use a resulting static set of rows as the static set of rows to load into the database buffer cache.

18. The database management system of claim 16, comprising the one or more computing devices further configured to:

generate a plan to execute the clause that includes;

performing a first database operation that produces a first static set of rows; and

after performing the first database operation, consuming the first static set of rows to perform a second database operation that produces a second static set of rows;

estimate a first computational cost of executing the plan by buffering the first static set of rows;

estimate a second computational cost of executing the plan by buffering the second static set of rows; and

generate a query execution plan based on the lowest estimated computational cost.

19. The database management system of claim 18, comprising the one or more computing devices further configured to:

generate a plurality of query execution plans that vary an order in which the database operations are executed.

20. The database management system of claim 16, wherein executing the query is performed by a plurality of database server instances;

wherein each database server instance has a respective database buffer cache of a plurality of database buffer caches;

wherein the database management system comprises the one or more computing devices further configured to;

distribute the static set of rows across the plurality of database buffer caches.

21. A method comprising:

receiving a query that requires an iterative execution of at least one sub-query within the query;

prior to executing the query,

identifying, based on the query, one or more static data inputs required for the iterative execution;

testing different permutations of a query execution plan by estimating a cost savings accrued due to grouping and buffering one or more of the static data inputs; and

selecting, for query execution, the query execution plan that maximizes the cost savings.