amazon simple storage

I recently learned about the Amazon Simple Storage Service (S3) and now I'm starting to move my backups into the cloud. Beforehand I had my own backup server running in the office and I would rsync each of production servers every night as well as take weekly snapshots on a month rotation. Now I can just put it all in the cloud and using s3cmd sync I can do the same sort of sync schedule.

Some things aren't quite as flexible as good old rsync though. I used to only sync files under 200MB so that I wouldn't bother backing up the odd linux distro iso that I'd leave on a server accidentally, or attempt to backup the virtual machine vdi files and I've had to work around that another way with s3.

Also I couldn't find a way to take a snapshot and copy that into another bucket with s3cmd, so every week I need to sync to the appropriate weekly snapshot buckets. This causes a bit more traffic than is really necessary but traffic between the data centre and amazon s3 is a lot cheaper than coming through my isp onto the office server.