[DRAFT] Reporting Stack

This page was created to document the tech stack of the current Reporting Stack.

Reporting Stack platform is made of multiple open source components, Apache Nifi, Apache Kafka, Druid and Apache SuperSet.

 

Superset
version: 0.29.0rc7

A description of this tool can be found here.

Importantly, Superset in OpenLMIS uses Superset-patchup. This is is a python package that "patches" Superset to add custom functionality that we find to be useful. You can find it here.

 

Kafka
version:

 

 

Druid
version:

 

Nifi
version: 1.9.2

NiFi is used for pulling data from OpenLMIS’s APIs, merging data from the APIs into a single schema, and transforming the data into a format that’s easy to query in Druid. There are currently two process group in NiFi: Generate Measures and Generate MeasureReports. These two process groups send data to the Hapifhir service. They run every day - 15 minutes after midnight.

 

Postgres
version: 9.6-debezium