1. I have data available in MySql, S3, Mongo, ElasticSearch.
2. We want to provide analytics on top of these data with different drilldowns, charting, dashboarding across the data sources.
3. Ability to discover the datasets from each data sources and create a semantic layer view across those datasets on which we can fire sql queries with projections, filters, groupbys, joins. Etc.
4. Custom report creation on top of these datasets and scheduled reports to send as excel csv files.
When we fire sql queries then the query actually gets executed against source db or against Druid. In that case whatever datasets we want to query we have to ingest all of that to Druid first.
Do you support joins over different dataets in druid ?
Can we make druid store data in S3 or we need to have the data in local disk.
Regards
Sambit