Monitoring database processes to generate machine learning predictions
AssigneeOracle International Corporation
InventorsSudhir Arthanat, Prashant Prakash
Life of this patent
Estimating the term…
Abstract
Methods and system are presented for monitoring database processes to generate machine learning predictions. A plurality of database processes executed on database implementations can be monitored, wherein the monitoring includes determining a start time, an end time, and a number of rows impacted by portions of the database processes, and the monitored database processes generate instances of machine learning data including at least the number of rows impacted and an associated duration of time. Using a machine learning component and the machine learning data, a duration of time can be predicted for a candidate database process for execution on a database implementation.
Claims20
A method for monitoring database processes to generate machine learning predictions, the method comprising:
- receiving a selection of structured query language (SQL) statements for monitoring;
- monitoring a plurality of database processes executed on database implementations, the monitored database processes comprising at least one of the selection of SQL statements, wherein the monitoring comprises determining a start time, an end time, and a number of rows impacted by the selection of SQL statements, and the monitored database processes generate instances of machine learning data comprising at least the number of rows impacted and an associated duration of time;
- and
- predicting, using a machine learning component and the machine learning data, a duration of time for a candidate database process for execution on a database implementation, wherein the machine learning component matches the candidate database process to one or more instances of the machine learning data generated by the monitoring to predict the duration of time for the candidate database process.
A system for monitoring database processes to generate machine learning predictions, the system comprising:
- a processor in communication with a storage device, wherein the processor is configured to:
- receive a selection of structured query language (SQL) statements for monitoring;
- monitor a plurality of database processes executed on database implementations, the monitored database processes comprising at least one of the selection of SQL statements, wherein the monitoring comprises determining a start time, an end time, and a number of rows impacted by the selection of SQL statements, and the monitored database processes generate instances of machine learning data comprising at least the number of rows impacted and an associated duration of time;
- and
- predict, using a machine learning component and the machine learning data, a duration of time for a candidate database process for execution on a database implementation, wherein the machine learning component matches the candidate database process to one or more instances of the machine learning data generated by the monitoring to predict the duration of time for the candidate database process.
A non-transitory computer readable medium having instructions stored thereon that, when executed by a processor, cause the processor to monitor database processes to generate machine learning predictions, wherein, when executed, the instructions cause the processor to:
- receive a selection of structured query language (SQL) statements for monitoring;
- monitor a plurality of database processes executed on database implementations, the monitored database processes comprising at least one of the selection of SQL statements, wherein the monitoring comprises determining a start time, an end time, and a number of rows impacted by the selection of SQL statements, and the monitored database processes generate instances of machine learning data comprising at least the number of rows impacted and an associated duration of time;
- and
- predict, using a machine learning component and the machine learning data, a duration of time for a candidate database process for execution on a database implementation, wherein the machine learning component matches the candidate database process to one or more instances of the machine learning data generated by the monitoring to predict the duration of time for the candidate database process.
Dependencies are inferred from the claim text.