US 12,067,009 B2
Predictive query parsing time and optimization
Bhashyam Ramesh, Secunderabad (IN); Jaiprakash Ganpatrao Chimanchode, PeeranCheruvu Hyderabad (IN); Naveen Thaliyil Sankaran, Kerala (IN); and Jitendra Yasaswi Bharadwaj Katta, Andhra Pradesh (IN)
Assigned to Teradata US, Inc., San Diego, CA (US)
Filed by Teradata US, Inc., Dayton, OH (US)
Filed on Dec. 10, 2018, as Appl. No. 16/214,280.
Prior Publication US 2020/0183936 A1, Jun. 11, 2020
Int. Cl. G06F 16/2453 (2019.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); H04L 41/5003 (2022.01)
CPC G06F 16/24545 (2019.01) [G06F 16/2453 (2019.01); G06N 3/08 (2013.01); G06N 20/00 (2019.01); H04L 41/5003 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method, comprising:
predicting a query parsing time using a trained machine-learning algorithm, wherein the predicted query parsing time comprises an estimated amount of time necessary for a query parser to fully parse a query, before the query parser fully parses the query, wherein the predicting comprises:
processing the query to extract features present in the query without fully parsing the query;
inputting the features to the trained machine-learning algorithm; and
receiving as an output from the trained machine-learning algorithm the predicted query parsing time; and
processing the predicted query parsing time in a query optimizer, the query optimizer using the predicted query parsing time as a factor in making different optimizations for the query, and producing a query execution plan for the query that accounts for the predicted query parsing time.