I would look into Git Annex which I have been heavily using to do the same thing with almost the same exact number of files. Its very simple to have a "dumb" d

I would look into Git Annex which I have been heavily using to do the same thing with almost the same exact number of files. Its very simple to have a "dumb" drive that simply stores the files and use annex to remember which drive it is. Also, to track and remember that you only have it in one place in case you want more copies. It can also push files to s3, glacier, and some other backup repos if needed with client side encryption (code between either symmetric or gpg)