Granted Patent B2

US 12,455,880 · App. 18/075,781 · Granted Oct 28, 2025

Inventors: Taehyung Lee (Seoul, KR); Taeyoung Jeong (Seoul, KR); Taehun Kim (Seoul, KR); Sungwon Kim (Seoul, KR); Sanghee Lee (Seoul, KR)

Assignee: SAP SE

G06F16/2454G06F16/24542

Abstract

Claims (52)

1 . A computer-implemented method for improving efficiency of parameterized query optimization, the method comprising:

receiving a parameterized query written in a declarative language, wherein the parameterized query comprises a placeholder which can be assigned different values when the parameterized query is executed;

generating a first query plan, wherein generating the first query plan comprises performing a first compilation session of the parameterized query in which the placeholder has no assigned value, wherein performing the first compilation session comprises:

generating an intermediate representation of the parameterized query, wherein the intermediate representation describes a relational algebra expression to implement the parameterized query; and

generating the first query plan based on the intermediate representation;

serializing the intermediate representation into an abstract query plan; and

generating a second query plan different from the first query plan, wherein generating the second query plan comprises performing a second compilation session of the parameterized query in which the placeholder has an assigned value, wherein performing the second compilation session comprises:

providing the abstract query plan and the assigned value as input to the second compilation session;

deserializing the abstract query plan into the intermediate representation; and

reusing the intermediate representation to generate the second query plan; and

executing the parameterized query using the second query plan.

2 . The method of claim 1 , wherein the intermediate representation represents the parameterized query as a query tree object comprising a plurality of nodes connected by edges, wherein the nodes represent database operations.

3 . The method of claim 1 , wherein the abstract query plan is written in a data exchange format that is transportable between different compilation sessions.

4 . The method of claim 1 , wherein one or more compilation steps of the first compilation session that are used to generate the intermediate representation are skipped in the second compilation session.

5 . The method of claim 4 , wherein the skipped one or more compilation steps operate independently of value assigned to the placeholder.

6 . The method of claim 4 , wherein the skipped one or more compilation steps comprise steps for parsing the parameterized query to generate an abstract syntax tree.

7 . The method of claim 6 , wherein the skipped one or more compilation steps comprise steps for normalizing the abstract syntax tree.

8 . The method of claim 6 , wherein the skipped one or more compilation steps comprise steps for converting the abstract syntax tree to the intermediate representation.

9 . The method of claim 8 , wherein converting the abstract syntax tree to the intermediate representation comprises sequentially applying one or more algebraic transformation rules to the abstract syntax tree.

10 . A computing system for improving efficiency of parameterized query optimization, the system comprising:

memory;

one or more hardware processors coupled to the memory; and

one or more computer readable storage media storing instructions that, when loaded into the memory, cause the one or more hardware processors to perform operations comprising:

receiving a parameterized query written in a declarative language, wherein the parameterized query comprises a placeholder which can be assigned different values when the parameterized query is executed;

generating a first query plan, wherein generating the first query plan comprises performing a first compilation session of the parameterized query in which the placeholder has no assigned value, wherein performing the first compilation session comprises:

generating an intermediate representation of the parameterized query, wherein the intermediate representation describes a relational algebra expression to implement the parameterized query; and

generating the first query plan based on the intermediate representation;

serializing the intermediate representation into an abstract query plan; and

generating a second query plan different from the first query plan, wherein generating the second query plan comprises performing a second compilation session of the parameterized query in which the placeholder has an assigned value, wherein performing the second compilation session comprises:

providing the abstract query plan and the assigned value as input to the second compilation session;

deserializing the abstract query plan into the intermediate representation; and

reusing the intermediate representation to generate the second query plan; and

executing the parameterized query using the second query plan.

11 . The system of claim 10 , wherein the intermediate representation represents the parameterized query as a query tree object comprising a plurality of nodes connected by edges, wherein the nodes represent database operations.

12 . The system of claim 10 , wherein one or more compilation steps of the first compilation session that are used to generate the intermediate representation are skipped in the second compilation session.

13 . The system of claim 12 , wherein the skipped one or more compilation steps operate independently of value assigned to the placeholder.

14 . The system of claim 12 , wherein the skipped one or more compilation steps comprise steps for parsing the parameterized query to generate an abstract syntax tree.

15 . The system of claim 14 , wherein the skipped one or more compilation steps comprise steps for normalizing the abstract syntax tree.

16 . The system of claim 14 , wherein the skipped one or more compilation steps comprise steps for converting the abstract syntax tree to the intermediate representation.

17 . The system of claim 16 , wherein converting the abstract syntax tree to the intermediate representation comprises sequentially applying one or more algebraic transformation rules to the abstract syntax tree.

18 . One or more non-transitory computer-readable media having encoded thereon computer-executable instructions causing one or more processors to perform a method for improving efficiency of parameterized query optimization, the method comprising:

receiving a parameterized query written in a declarative language, wherein the parameterized query comprises a placeholder which can be assigned different values when the parameterized query is executed;

generating a first query plan, wherein generating the first query plan comprises performing a first compilation session of the parameterized query in which the placeholder has no assigned value, wherein performing the first compilation session comprises:

generating an intermediate representation of the parameterized query, wherein the intermediate representation describes a relational algebra expression to implement the parameterized query; and

generating the first query plan based on the intermediate representation;

serializing the intermediate representation into an abstract query plan;

generating a second query plan different from the first query plan, wherein generating the second query plan comprises performing a second compilation session of the parameterized query in which the placeholder has an assigned value;

wherein performing the second compilation session comprises:

providing the abstract query plan and the assigned value as input to the second compilation session;

deserializing the abstract query plan into the intermediate representation, and

reusing the intermediate representation to generate the second query plan; and

executing the parameterized query using the second query plan.

Assignments (1)

Continuity (2)

References Cited (24)