We are building Tableau dashboards based on Snowflake and is using live-connection to take full advantage of the Snowflake architecture.
The dashboard has multiple sheets which connects to the same data source. When the dashboard loads, we observe that each sheet fires query to the same data source which is turning out to be costly.
Questions:
1) Is there a way to fire a single query for the above dashboard since it uses the same data source for all the sheets?
2) Are we able to use any cache mechanism to accomplish above? This will help query data source once and use the cache to load all the charts.
3) What are the best practices to optimize dashboards based on live connection when using DBs like snowflake?
Your first post implies that you're using a live connection and trying to reduce the Snowflake costs - so extracts is an option.
Now you say you're using extracts and looking to move back to a live connection - that will increase costs.
Snowpipe won't fix that problem at all.
Check the load on your warehouse, you might be able to use something smaller (reducing costs). Also how are you maintaining extracts at the moment ? You should be able to schedule in Server/Online and it's a "set and forget" exercise.
