A complete solution for open data platforms, enterprise data catalogs, data lakes and data management. Open source, mature, fully-featured and production ready.

DataHub

Screen Shot 2021-08-20 at 3.15.35 PM.png

Have the following Questions :
1. How to use a different Database for GMS service, any documentation that we can follow ?
2. To run Datahub, both Graph Index  and Search are required ? What are used for Graph Index and Search Index by default ?Referring to the Architecture diagram shared

<@U029VFMT1GV>: by default we use Elastic for both Search and Graph index currently.

<@UV0M2EB8Q> - What is the role of neo4j ?
Also any direction on the first question is greatly appreciated. ie
1. How to use a different Database for GMS service, any documentation that we can follow ?


You can use neo4j for graph index as well if you want. multi-hop queries are more efficient in neo4j ... however the UI currently doesn't use them, so there isn't a strong reason for using it, unless you want to query the metadata programmatically

<@U029VFMT1GV> For the GMS service -&gt; database question: there are community members using MySQL / Postgres / MariaDB / AWS RDS etc. Any thing that works with EBean works. A quick skim over the docs doesn't turn up any results for how to configure different databases though. Which database were you planning to use?

<@U01NN9CTQ01>: what are the steps for connecting datahub-gms to postgres?

Take a look at the docker env here <https://github.com/linkedin/datahub/blob/master/docker/datahub-gms/env/docker.postgres.env#L3>. You need to set EBEAN_DATASOURCE_URL to point to postgres instead of mysql and change the EBEAN_DATASOURCE_DRIVER

<@U01NN9CTQ01> - If we were to modify the above is that sufficient to connect to a different DB (not running as a docker ), but a standalone DB.