What protocols do you have available for opening the file?
A 55G file doesn't sound too bad, can you just download it and handle it locally?
Clearly it would be better if the data providers made more smaller files; a very large file is difficult to handle.
I might suggest using rsync (if you can) as that will enable resuming if the transfer fails (it's going to take a couple of hours I guess, if it's from a remote site).
How often do these files get generated? Can you get the party generating them to create deltas instead? That would probably be a better way to handle it (assuming the data aren't entirely new each time)