We currently have an enterprise geodatabase that we use for hosting read-only web services. I'm wondering if this would be better suited for a local file geodatabase on our ArcGIS Server machine, in terms of read performance?
What I'm really interested in is: is there a way to get quantifiable data that would show performance differences? For example, if I had the same feature class in an enterprise gdb and a file gdb, and each are published to separate services, can I be able to determine any noticeable performance gains? ... are the ArcGIS Server statistics a good method for doing this?...are there other tools that would help detect performance differences?
My initial thoughts would be: a local file gdb should perform better since it's local to the Server machine, and since it's a file gdb, it doesn't have to authenticate a user, read any versions, and connect to a different server.
I'm not sure if it would be leaps and bounds better in performance, and maybe it wouldn't provide any noticeable performance gains at all...I'm just curious on if others have tried detecting performance differences, and if so, what methods, tools, etc... did you use and which were helpful.