If this is what you are talking about, I would have a good read through it and its subsection to see if there is any advantage or disadvantage to your proposal
We have a clustered environment that all point to a file share. There is a shared folder on the file share that houses the config store. Also, in that folder there separate folders for mxds and data. We are trying to figure out if this causes any operational issues for the future to have one shared folder but separate folders for to the config store, separate for the mxds, and separate for the data.
That is the type of question I would ask tech support, unless someone wades in with an exact duplicate of your desired situation. One 'nice' could lead to a whole load of trouble if it isn't advisable and tech support has probably seen more installation conditions than anyone else. My advise... ask the pros
I'd agree with Dan. I can say we also use a shared UNC location for the config store, and server job folders, etc. I keep a backup copy of all of our DB connection files, ETL's etc. in a folder at the same location but it's not the true source for any of our work. Thus far it's never caused any issues.
From an architectural stand point, I only see 2 downsides. First, you create a single point of failure. So if you are backing up this share location in its entirety elsewhere you have a restore option for not just servers but data, etc. as well. The only other concern you run is depending on what/how you are leveraging server and the data for desktop needs etc. in simultaneous workflows because you could push up against some bad disc I/O giving performance headaches to users on both ends (desktop & server).