Recently we started hitting into a situation where we are simply putting more into the queue than can be "picked up" by the data engine. Our main item doing this is the SQL databases where it returns a lot of historical metrics, even with most of the QoS options turned off. I've looked for information on this but haven't seen an answer to my question. Is it possible to have two or more data engines running to the NimBUS database? I am curious if I can have one handle the SQL metrics, one for system type metrics, etc. Has anyone else run into a problem where they are consistently having the data_engine queue fill up and not empty out?