Systems and methods for cardinality estimation feedback loops in query processing

Methods for cardinality estimation feedback loops in query processing are performed by systems and devices. The query host executes a query for the data source via the engine based on the estimated cardinality, and the query monitor generates an event signal during execution and upon completion. The...

Full description

Saved in:
Bibliographic Details
Main Authors LOPEZ PAULO M, MAHAJAN AMIT, GEDZIOGZEWSKI, GRZEGORZ, CHOI IN-JIN, SACK, JAMES, I, REDMAN, JOHN, L., JR, EXARRE VISCHIA, ANTONIO, KADIAM CHARLOTTE, PAPADIMOS, VASILIOS, XING NAN
Format Patent
LanguageChinese
English
Published 01.04.2022
Subjects
CALCULATING
COMPUTING
COUNTING
ELECTRIC DIGITAL DATA PROCESSING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:Methods for cardinality estimation feedback loops in query processing are performed by systems and devices. The query host executes a query for the data source via the engine based on the estimated cardinality, and the query monitor generates an event signal during execution and upon completion. The event signal includes actual data radix, runtime statistics, and indicia of query parameters in the query plan, and is routed to an analyzer of the feedback optimizer, where the event signal information is analyzed. The feedback optimizer utilizes the analysis results to generate change recommendations as feedback for subsequent execution of queries or similar queries executed by the query optimizer of the query host. The query host stores change recommendations and monitors subsequent queries for the same or similar query, the change recommendations for the query being applied to the query plan for execution and observation by the query monitor. The change recommendations are selectively viewed and selected via a
Bibliography:Application Number: CN202080040037