Overview
IBM Z® Resource Discovery Data Service (ZRD-DS) is an embedded component in several IBM® products. It is used to discover the infrastructure componentry of a system, understand the relationships and interdependences between the components, and connect these artifacts to a topological UI via a graph API. It can help you perform root-cause analysis faster by understanding how artifacts impact one another based on the discovered relationships. Although integrations and use cases are continuously expanding, examples of ZRD-DS consuming products include:
- IBM Wazi Analyze inside IBM Application Discovery and Delivery Intelligence for IBM Z which allows users to understand the impact of making application code changes.
- IBM Cloud Pak® for Watson™ AIOps which enables events that originate from IBM OMEGAMON® monitoring tools or System Automation to be correlated with events from hybrid cloud environments.
- IBM Z Anomaly Analytics with Watson which allows users to understand the relationships that are associated with anomalous artifacts, and assess impact or find the root cause of issues faster.
Architecture
As illustrated in the following diagram, IBM Z Resource
Discovery Data Service includes the following basic components: 
- Discovery Data Core Service
- Discovery Data Core Service is for managing resource data, defining the data sources and performing ETL processes to generate Discovery data in to a graph database.
- Discovery Data API Service
- A RESTful API which provides data query in JSON Graph Format.
- Discovery Data Store
- Discovery Data Store is for storing data discovered. Discovery Data Store includes two databases, PostgreSQL and OrientDB. PostgreSQL stores raw resources data that is discovered and generated by Discovery Data Core Service that transforms the data to Discovery relational data model. OrientDB is a graph database for storing resource objects and relationships. Discovery Data API service reads data from OrientDB and provides JSON Graph format data via RESTful API.
- PostgreSQL Kafka Connect
- PostgreSQL Kafka Connect is a separate container image which obtains a snapshot of the existing data in IBM Z Resource Discovery Data Service PostgreSQL database and then monitors and records all subsequent row-level changes that are committed to the data. It uses the Kafka Connect API to register itself as one of the connectors after the changed data is available in an understandable format and converts changes to event streams. All of the events for each table are recorded in a separate Apache Kafka topic. The events can then be easily consumed by applications and services.