Large Database file

[Eric Aiken]

I know this is an open-ended and loaded question, I'm looking forward to thought provoking comments.

I need to load into a clean dB/server 78 Million address records (National Address Database) with a csv, version of the file. I have the option of loading a gdb version (geospacial db). Not knowing anything about gdb files/directories, I didn't want to add to the complications on such a large file.

Any thoughts about where to begin?
break the db files up in any particular way?
Memory settings?
other dB settings.

Each row is relatively small, just lots of them.
I'm not going to do much with it (no production use), other than build some queries and export portions of it.

It'll be running off a desktop, so I don't expect performance. I've already watched someone's video on the challenges of loading it. There are some field issues with respect to the defined schema.

Thanks in advance for any thoughts, feedback and suggestions.