Gitlab 10 and S3 Object Storage

Good Morning - We are looking into the possibility of upgrading our install to the enterprise edition of gitlab which apparently supports Object Storage using S3 as the backend. This is intriguing for us as we often run into issues where we want to store large files and have gitlab manage them. According to the documentation “Object storage currently transfers files first to GitLab, and then on the object storage in a second stage.” Does anyone know if this would work on a 40gig file? Or, does anyone know if its possible to first upload the file to S3 and then make gitlab aware of it and able to manage it?

This could be the exact feature we need in order to deal with these rogue large files. Any input is greatly appreciated.

Thank you!