US 12,361,002 B2
High-performance large object operations
Shen Yan Sun, Beijing (CN); Peng Hui Jiang, Beijing (CN); Xiao Xiao Chen, Beijing (CN); and Xiao Hui Wang, Beijing (CN)
Assigned to International Business Machines Corporation, Armonk, NY (US)
Filed by INTERNATIONAL BUSINESS MACHINES CORPORATION, Armonk, NY (US)
Filed on Sep. 11, 2023, as Appl. No. 18/464,753.
Prior Publication US 2025/0086181 A1, Mar. 13, 2025
Int. Cl. G06F 16/30 (2019.01); G06F 16/22 (2019.01); G06F 16/2453 (2019.01)
CPC G06F 16/24549 (2019.01) [G06F 16/2219 (2019.01); G06F 16/24542 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A computer-implemented method of planning and executing a query referencing a large object (LOB) in a database system, the method comprising:
training, by one or more processors, a classifier, the training comprising:
building, by the one or more processors, an implicit column usage knowledge base, the building comprising:
obtaining, by the one or more processors, sample data comprising query strings referencing the LOB for different usages and query strings that do not reference the LOB;
iteratively executing, by the one or more processors, the query strings as a background processes;
outputting, by the one or more processors, based on the iteratively executing, query string results comprising the performance benchmarks; and
retaining, by the one or more processors, the query string results in the implicit column usage knowledge base; and
applying, by the one or more processors, at least one machine learning algorithm to the implicit column usage knowledge base to train the classifier, wherein based on the machine learning, the classifier is the trained classifier;
obtaining, by the one or more processors, a given query for execution in the database system, wherein the given query comprises at least one reference to the LOB;
obtaining, by the one or more processors, statistics and activities performed in the database system;
establishing, by the one or more processors, performance benchmarks for executing queries in the database system based on analyzing the statistics and the activities to set performance;
analyzing, by the one or more processors, the given query to determine how the LOB in the given query is utilized in the given query;
determining, by the one or more processors, based on at least one performance benchmark of the performance benchmarks and based on the analyzing of the given query, that meeting the at least one performance benchmark requires replacing the at least one reference to the LOB with a reference to one or more implicit columns, wherein the determining that meeting the at least one performance benchmark requires replacing the at least one reference to the LOB with the reference to one or more implicit columns comprises applying the trained classifier for the determining;
executing, by the one or more processors, the given query, wherein the executing comprises the given query referencing the one or more implicit columns to obtain query results for the given query; and
returning, by the one or more processors, the query results.