Dear all, I want to connect to a single MonetDB server from an application using the mapi interface. The intention is to load a small input dataset using insert statements and mapi_query(). Then I will execute a SELECT statement from mapi against another dataset resident in the MonetDB server and return the results to the application. My question is about possible concurrent connections from this application. Is it sufficient to use unique table names for my input data sets so each input data set ends up in a separate table? Is there a better approach for this use case built in to MonetDB that I have missed? Best regards, Alastair