About
Riffl is a generic streaming data ingestion framework currently built on top of Flink and leveraging Table API.
It aims for its process to be simple to define and reason about with Yaml configuration and SQL expressions. Deploys into wide range of environments be it Hadoop, Kubernetes or in any other Flink supported ways and it is self-contained.
Riffl puts data quality first with exactly-once guarantees but also output optimization so that query engines can utilise their features to operate more efficiently.