Hi there, may I know what’s the frequency of Druid Lookup will pull data from RDMBS database (SQL Server)?
The reason why I am asking this is that I need to make sure that the lookup is not creating heavy load for the SQL Server side.
Also, is this configurable in the lookup config?
Another question, is it recommended to add indexing to the columns that need to be lookup by Druid to help with the performance issue on the rdbms server?
This is an example of a lookup that is being polled every 10 minutes. It is configurable.
It would depend on the database, but usually indexes speed up query processing in most RDBMS
Thanks for the reply.
The “Globally cached lookup type” that I am using is through JDBC, there’s no “Poll period” available on the Druid user interface.
JDBC, doesn’t have poll period option https://i.imgur.com/WdTOzyK.png
uri method, have poll period option https://i.imgur.com/uTSNU2i.png
I have also tried to create the lookup via Druid API with the configuration below: https://i.imgur.com/LsZD9Py.png But it doesn’t seem that the data polling is following that as I can’t see the data in lookup is being updated even after I updated the data in SQL Server.
May I know if lookup through JDBC (SQL Server) support “poll period” configuration?