Now that we have a digital camera, the collection of digital photos on our home computer is growing in leaps and bounds. And as I mentioned before, we have a bunch of home videos as well. If there were a disk crash or a fire or some other disaster, we’d lose everything, so I started looking into backup strategies. A friend of Gail’s bought a big USB hard drive and backs everything up to that, but unless you store the drive somewhere else and bring it home occasionally for backups, a fire would destroy both the original and the backup, so that’s no good. I could just burn DVDs and then take them to work or something (putting them in the safety deposit box would be the best solution, but I’m way too lazy to do that) but that’s a lot of manual labour, and I’d need numerous DVDs to store everything. Too much work, and so I know that I wouldn’t do it often enough. What I need is something that does automated backups online so that the files are stored offsite, and the backups can be done automatically so I don’t need to remember to do it.
I think I’ve found the solution thanks to the Security Now! podcast. Steve Gibson talked about a tool called Jungle Disk, which uses Amazon’s S3 storage service. It’s very cheap — about 15 cents per gigabyte per month, you only pay for what you use, and there are no storage limits. You also pay 10 cents per gigabyte uploaded, and 18 cents per gigabyte downloaded. The Jungle Disk software itself is only $20, comes with free lifetime upgrades and you can install it on multiple machines. Once it’s installed and you’re hooked up to your Amazon account, Jungle Disk sets up a virtual drive on your Windows machine, so you can copy files to and from the “disk” at will, so if you want, you can essentially use it as a huge USB key (though remember the upload/download fees). It also has fully-functional automated backup software that has strong 256-bit AES encryption (which is why it was mentioned on a security podcast).
The one thing it doesn’t do is compress the data. This is supposedly because of a feature they have called block-level updates, where they backup parts of a file individually, so that if your 2GB file changes now and again but only 50 MB of the data is actually changed, it will back up just the changed parts rather than the entire file each time, saving bandwidth. Compressing a file would break this, and it’s probably true that the block-level updates save you money in the long run for large frequently-changing files, but for files that never change (like my home videos), compression would be way more useful. Plus, the block-level update feature is only available if you purchase the optional add-on service called “Jungle Disk Plus”. Maybe I’ll have to set up a cron job (or Windows equivalent) that runs once a week (the night before the backup, probably) and compresses the videos.
The ability to use the Jungle Disk service as a huge USB key is rather nice as well — if I want to transfer a file from work to home, I can just copy it up to my Jungle Disk drive at work, then copy it down at home. I’d have to pay the upload/download costs for that, but depending on the file, it may be worth it. When I’m at home, both my work and home machines use the same wireless router to connect to the internet, but neither can see the other’s shares. I don’t know why I can’t connect them to each other, and neither Gail nor I have any idea how to fix it, though admittedly our research into the problem has been minimal. In the past when I wanted to transfer a file from one machine to the other, I either used a regular USB key or burned a DVD for larger files. This gives me another option.
I wrote this article several weeks ago but never got around to posting it. I just got my first bill from Amazon Web Services for all of $1.87. I don’t quite understand it though — it says I’m using 5.305 GB of storage, but uploaded 9.217 GB of data. It backed up all of our France pictures this past weekend, so the next bill will be much higher, but we’ll see how much more it is.