Need some input, file manipulation, backup

[Brynna Childs]

Hi I work at a media arts organization with a complex (enough) file-saving structure, and multiple departments, projects and network drives. I am newly appointed to create a DAM system in ResourceSpace. 

We are testing some workflow things and noticed that after import to RS, our files became gigantic! Anyone know why? 

I also wondered if a copy on import is always necessary (or best) and if it's possible to work of a NAST drive directly from a linux server and reduce or eliminate the need to make a working copy. Given the ability to revert in RS, and our additional lock-box online automized backup, I see no reason to copy the file on import (in this case on our designate linux desktop/server) to then export a final backup moments later. 

Anyone have a few minutes to answer some of our primary questions? I am an archivist,  but this is my first run with a DAM (so far, it's pretty fun, but we're testing feasibility namely because we are working with large file sizes, temporary storage seems to be an obstacle). 

The team I am working with is a mix of student and community workshop participants with varied experience levels in digital media, very professional and experienced producers, designers, and academics, some working internationally, and desktop publishing-forward research assistants. 

We want to use the DAM primarily to encourage internal file access, and streamline metadata applications most primarily for searchability across drives, projects, with imperfect/ not always consistent naming conventions. 

Thanks and hello.