gRPC Conf 2019

Protobuf and gRPC are build from the ground up to be forward and backward compatible. Due to this compatibility the need of a native schema registry has never got off the ground. Sharing was simple done by sharing git repositories.

In the big data ecosystem Kafka is gaining popularity and the schema registry is a centerpiece of architectures build on top of Kafka, while proto doesn’t have an answer to this. That's why we build metastore, a schema registry, to facilitates a contract first workflow for gRPC and Protobuf across different teams. The store is able to collect, validate and even lint proto contracts against historical version but still backed by a git repo.

This talk will present metastore and a few workflows that uses the store. But we’ll also try to propose a standard gRPC that could be part of the gRPC spec.