US 11,720,535 B2
Data migration and integration system
Eric Alexander Daimler, San Francisco, CA (US); Ryan Jacob Wisnesky, San Francisco, CA (US); and David Isaac Spivak, Somerville, MA (US)
Assigned to Conexus ai, Inc., San Francisco, CA (US)
Filed by Conexus ai, Inc., San Francisco, CA (US)
Filed on Jan. 10, 2022, as Appl. No. 17/571,939.
Application 17/571,939 is a continuation of application No. 16/844,810, filed on Apr. 9, 2020, granted, now 11,256,672.
Claims priority of provisional application 62/832,214, filed on Apr. 10, 2019.
Prior Publication US 2022/0129425 A1, Apr. 28, 2022
Int. Cl. G06F 16/21 (2019.01); G06F 16/84 (2019.01); G06F 16/182 (2019.01)
CPC G06F 16/214 (2019.01) [G06F 16/182 (2019.01); G06F 16/211 (2019.01); G06F 16/86 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A system, comprising:
a memory configured to store a mapping from a source schema of a source database comprising source data to a target schema of a target database to be populated with desired target data, wherein the mapping comprises a left-Kan extension relationship between the source data and the desired target data; and
a processor coupled to the memory and configured to migrate to the target schema a database instance of the source schema, wherein the migrating, based on left-Kan extensions via the mapping to transform the source data, is implemented at least in part by a chase engine executed on a computer to invoke a chase algorithm to perform an ordered sequence of steps comprising:
adding, based on one or more data integrity constraints of the target schema, a bounded layer of new elements to the target data after a chase state has been initialized, wherein the chase state iteratively approximates a desired left-Kan extension of the source data according to the data integrity constraints in the target schema;
adding coincidences in the chase state based on the data integrity constraints of the target schema and the mapping from the source schema to the target schema, wherein the coincidences are in part induced by a functoriality that maps equal elements between the source schema to the target schema via the mapping;
merging the equal elements based on the coincidences to output a final chase state that is uniquely isomorphic to the desired left-Kan extension, wherein the final chase state additionally contains a universal property of the desired left-Kan extension, the universal property comprising a data mapping from the source data to target data; and
repeating the ordered sequence of steps iteratively until an end condition is met, wherein the end condition is based at least in part on a determination that no further new coincidences or new elements exist to be added.