PI tag configuation for a RDBMS

another option may be to change your SQL query to return a single row with the largest timestamp?

If you are only going to process that one anyway, it could be better for the network and other resources not to return a bunch of unuseful rows

another option may be to change your SQL query to return a single row with the largest timestamp?

If you are only going to process that one anyway, it could be better for the network and other resources not to return a bunch of unuseful rows

Well this is only a suggestion and not meant to be a direct solution to rbitcoin​'s desire (although a slight change should do). I do agree with you that the query result should be as optimized and light-weighted as possible, but from personal experience getting SQL data from external applications, human errors and sometimes negligence can require data to be adjusted and thus, require overriding PI data.

sahilp​ I like this approach. I'm thinking that there will be 2 queries. 1 query to get results for distribution and 1 query to get the latest timestamp. The query to get the data will use the largest timestamp in the where clause to fetch the latest data. Is this possible to do this all in one query and then update the latest timestamp accordingly after?