Transfer files from Google Drive to Amazon S3

Server to server, without downloading anything to your computer.

How it works

Connect both accounts once. ShiftMyData then talks to Google Drive and Amazon S3 directly and streams the files between them in memory — nothing is written to our disks, and nothing travels down to your machine and back up again. The job runs on a worker, so closing the tab does not stop it.

Will every file be verified?

Both sides expose MD5, so every file is compared by checksum after it lands. A file that arrived corrupted or truncated is detected rather than assumed good, and an unchanged file is skipped on the next run without re-reading it.

Before you start

The largest single file Amazon S3 accepts is 5 TB. A bigger one is reported as failed and the rest of the transfer carries on.
Google Drive allows two files with exactly the same name in one folder, and Amazon S3 does not. Only one of each such pair is copied; rename duplicates before you start if you need both.
Amazon S3 has no real folders, only names with slashes in them, so empty folders are not created there. Every folder that holds a file arrives as expected.
Google Docs, Sheets and Slides arrive as Word, Excel and PowerPoint files. Files shared with you and shared drives are separate roots in the explorer, not part of My Drive: copy them as well if you want them.
A transfer copies each file's current content, name and modification time; sharing and version history stay behind on Google Drive. Business migration carries them where both sides support them.

Sources: support.google.com · rclone.org · docs.aws.amazon.com · aws.amazon.com

Worth knowing about Google Drive

Google Docs, Sheets and Slides are not files in the ordinary sense -- they have no bytes to copy and no size until they are exported. ShiftMyData exports them to their Office equivalents on the way out; a native Doc cannot be preserved as a Doc anywhere except another Drive.

Worth knowing about Amazon S3

S3 has no modification time of its own -- it records when an object was written, not when the file was authored, and that timestamp cannot be set. Transfers into S3 therefore compare on checksum where one is shared, and on size where one is not.

Steps

On the Clouds screen, add Google Drive: sign in with Google and allow access.
Add Amazon S3 the same way: paste an access key ID and secret access key from the AWS console, ideally for an IAM user limited to the buckets involved.
Pick the source folder in the left pane and the destination bucket in the right.
Choose a mode — copy, mirror, move, or two-way — and start the transfer.
Leave it. Progress is on the Jobs screen, and a webhook can tell you when it ends.

Questions

Do I have to download my files from Google Drive first?

No. The transfer is streamed from Google Drive to Amazon S3 on our servers, so nothing is written to your machine and your home upload speed is not the limit.

What happens if I close my laptop?

The job keeps running. It is queued on a worker rather than in your browser, and the progress is there when you come back.

How do I know every file arrived?

Is there a size limit?

None imposed by us. What you move is drawn from your prepaid allowance (30 GB free, then 500 GB for ₹100 or 2,500 GB for ₹500, each valid a year). Amazon S3 takes single files up to 5 TB, and its own storage quota applies.

Move a folder now