Using Semantic Metadata to Create an Automated Microservice Data Mesh
summary
Claim: embedding Taxi semantic types into API/Schema metadata lets tooling build an on‑demand integration graph and answer business queries without prewired flows. Implementation uses OpenAPI/Protobuf/DDL plus a graph engine called Vine to stitch Postgres, REST, SOAP and Kafka sources; demo shows a film catalog lookup that hits three systems and returns reviews and streaming info. Consumer requests use a consumer contract (requested field shape), and when IDs or endpoints change the mesh reroutes and adapts automatically.
AI-written summary · about
chapters (10)
- 0:00Intro
- 0:38Data mesh
- 1:48Automated
- 4:40Semantic metadata
- 6:32Structural vs semantic contracts
- 19:50Automation
- 22:44Consuming using semantic metadata
- 24:38Putting it all together
- 24:56Demo
- 34:51Outro
related talks
57mOur DataMesh Journey: data architectures for microservices
37mMetadata Automation in the Data Mesh
46mJourney of metadata
15mEasy Metadata Management (Git for Data) Using Neo4j
29mMetadata Systems for End-to-End Data & Machine Learning
51mWhat is a Data mesh?
26mBuilding Scalable Microservice APIs From The Beginning
28mOntology Creation