How to: Backup Automatically to Amazon S3

Applies to Windows NT, XP and Vista.

Regular daily, weekly and monthly backups should be implemented in any business or situation where you can’t afford to loose items stored on your PC and servers. In the past this could be quite a costly process with tape drives and complex backup software.

However, with the drop in price of hard drives believed to fall by at least a 1/3 every year, the cost of an external hard drive to backup your whole system should now seriously be considered in any business no matter how small. Used in conjunction with excellent open source or freeware backup software available, comprehensive backups can be configured quickly, easily and automatically to your external hard disk.

I am gonna show you how to go one stage further though and backup your vital documents off site with Amazon S3. From Wikipedia:

Amazon S3 (Simple Storage Service) is an online storage web service offered by Amazon Web Services.

S3 launched back in 2006 and offers very competitive rates for file storage. At time of writing S3 cost $0.150 per GB – first 50 TB / month of storage used and $0.100 per GB – all data transfer in. Full pricing can be found here. In short their pricing is extremely competitive.

We are gonna use a bit of beta software called S3Drive which is still in development and requires the .Net framwork.

Backup to S3

All the usual caveats apply, make sure you have everything backed up before proceeding. I can not be held responsible for anything that goes wrong with these instructions.

  1. You must have an Amazon S3 account, you can sign up here
  2. Download S3Drive from here
  3. Run the S3Drive installation program (S3Drive0.6.msi), if you do not have .Net installed it should prompt you to install it but alternatively it is available from here
  4. Once installed you will need to configure your S3Drive. Run the ‘Configure S3Drive’ option from the start menu. configure
  5. Click on the first option, ‘Enter or Change your Amazon S3 access information’. You will need to enter your ‘Access Key ID’ and your ‘Secret Access Key’, both found in the Access Identifiers section when you log into your S3 account in your browser.
  6. Click on second option, ‘Define your Hard Disk….’. Enter a name for the bucket in the box, create the disk and then assign it a drive letter.new_disk
  7. Click Finish. The service will then be started

This should create a network share for your amazon bucket and assign the drive letter you set for it. However on first installation I had to manually map the network drive.

network-drive1If you navigate to My Computer / Computer now you should see this drive which works in the same way as any other hard disk attached to your computer except it is actually located on Amazon’s servers. Try copying a file to it to ensure all is working.

Next it is simple a case of configuring your software to backup to this ‘drive’. I am using Syncback to do this and my other backups. An excellent introduction and tutorial on configuring Syncback Freeware is available at Lifehacker.

There are a couple of  thing to bear in mind, firstly, you are backing up across the web. Unless you have an exceptional connection, backing up your mp3 collection is gonna take days and days. For large amounts of data, I would recommend splitting up and backing up in much smaller sizes. Secondly, this is not the complete solution, this should be run in conjunction with another backup solution too. In my case I back up on a daily, weekly and monthly basis to an external drive in addition to my really vital commercial work stored on S3.

There are paid alternatives to this such as JungleDisk which is available on Mac, Linux and PC. Further, my Mac buddies tell me the excellent ftp client, Transmit,  now offers S3 integration.

Good luck and if you have any questions I will do my best to help.

Leave a Reply