The big data matching includes
Java and REST interfaces and classes for searching and comparing your
data.
The big data matching provides
an API both as Java™ and REST.
The APIs do not overlap with the InfoSphere® MDM
APIs.
You can start endpoint coprocessors at any time from the
client using either of the
big data matching APIs.
Endpoint implementations are run remotely at the target region or
regions, and the results are returned to the client. The endpoint
coprocessors:
- An entity endpoint, which gets the entity ID (EID) for a record
- A search endpoint, which runs a probabilistic search for a record
or entity
- A compare endpoint, which gets a detailed comparison score between
two records
You can call endpoints in either of two ways:
- Call the endpoint directly through the Java API.
- Call the endpoint with the REST API using certain URLs.
The installer for the big data matching registers
a set of custom REST services automatically with the HBase REST server.
The services don't require additional configuration. You can call
them remotely by specifying the correct URL syntax and the format
of the results.
IBM InfoSphere Master Data Management, Version 11.3