Data drift detection between data storage
AssigneeIntuit Inc.
InventorsRaymond Chan, Suresh Muthu
Life of this patent
Estimating the term…
Abstract
A method for detecting data drift between a first database and a second database involves obtaining (from the first database) and based on a change data capture (CDC) event generated in response to a change detected in the first database, a first record identified by the CDC event, obtaining (from the second database) a second record corresponding to the first record, transforming a data structure of the first record from the first database to the data structure of the second database generating a transformed record, and based on determining that a difference between the first record and a second record exists, reporting a presence of data drift.
Claims20
A method for detecting data drift between a first database and a second database, comprising:
- obtaining, from the first database, and based on a change data capture (CDC) event generated in response to a change detected in the first database, a first record identified by the CDC event;
- obtaining, from the second database, a second record corresponding to the first record;
- obtaining a remapped first record by mapping the first record from a first data model of the first database to a second data model of the second database;
- comparing the remapped first record to the second record;
- determining, based on comparing, that a data drift exists, wherein the data drift comprises a difference between the first record and the second record;
- and
- mitigating the data drift by transforming a data structure of the first record from the first database to the data structure of the second database to generate a transformed record.
A system for detecting data drift detection between a first database and a second database, comprising:
- a computer processor;
- and
- a data drift detection engine executing on the computer processor configured to:
- obtain, from the first database, and based on a change data capture (CDC) event generated in response to a change detected in the first database, a first record identified by the CDC event;
- obtain, from the second database, a second record corresponding to the first record;
- obtain a remapped first record by mapping the first record from a first data model of the first database to a second data model of the second database;
- compare the remapped first record to the second record;
- determine, based on comparing, that a data drift exists, wherein the data drift comprises a difference between the first record and the second record;
- and
- mitigating the data drift by transforming a data structure of the first record from the first database to the data structure of the second database to generate a transformed record.
A non-transitory computer readable medium comprising instruction for execution on a computer processor to perform:
- obtaining, from a first database, and based on a change data capture (CDC) event generated in response to a change detected in the first database, a first record identified by the CDC event;
- obtaining, from a second database, a second record corresponding to the first record;
- obtaining a remapped first record by mapping the first record from a first data model of the first database to a second data model of the second database;
- comparing the remapped first record to the second record;
- determining, based on comparing, that a data drift exists, wherein the data drift comprises a difference between the first record and the second record;
- and
- mitigating the data drift by transforming a data structure of the first record from the first database to the data structure of the second database to generate a transformed record.
Dependencies are inferred from the claim text.