User-1426443870 posted
We have an internal web application that works just fine when using a single server. The application averages over 100,000 SignalR messages per day. However, for various reasons, we now need to deploy the application on multiple servers behind a load
balancer and would like to choose SQL Server as our backplane solution.
With this type of volume, we are concerned about how large the database will grow. In our development environment, we have found that truncating the tables does not work - the only way we have managed to keep the SignalR database small is to stop the application,
drop the tables and start the application again. This is not exactly a best practice for our production environment and we would like some recommendation as to how to keep the SignalR production database from experiencing exponential growth.
Also, we have set the tablecount parameter to 2 in our development environment. It would be nice to not have to experiment with what the optimal number of tables should be in our production environment.
Thanks and Regards,
Jim