Archiving data to the cloud

Computing “in the cloud” is where it’s obviously at. Amazon’s S3 service gives access to remote disk space which can be accessed via API’s; ditto EC2 for actually doing the processing. And then of course there’s numerous applications from Google Docs to Slideshare to web based email. Not to mention web based photo editors such as PXN8 and the rumoured online version of Adobe Photoshop. As connection speeds get faster, the vision of the internet-based thin client becomes reality.

All that is well and good. But there’s also a huge mind-wrangle which many institutions are still struggling with. Getting rid of Microsoft Exchange and replacing it with, say Google Apps as the mail/calendering/IM solution for your business is probably a no-brainer financially, but it’s a conceptual leap which many IT teams have real problems with. The same applies for storage. Off-hosting to an ISP is one thing; chucking all your assets into the cloud with a solution such as Amazon is another thing entirely.

Backup and archiving is a big issue for anyone, from museums to home computer users with a disk full of family images. After a couple of really horrible laptop moments, I learnt my lesson and now tend to back up everything I do to an external hard drive on a fairly regular basis. I still have the fear that these – literally priceless – assets (pictures of my son when he’d just been born, my gran just before she died, our wedding) could potentially disappear in the event of a hard drive meltdown or fire.

Sending data to the web for backup is the obvious solution. And although rationally I know that a hard drive sitting under my desk is more likely to get nicked, broken, burnt or eaten by a passing 3 year-old, I still really struggle with the idea that my files are out there somewhere. It’s a basic human response – if I can see my stuff (or at least the casing that surrounds it) then I know that everything is ok. If it’s somewhere on a random server (albeit one which is probably backed up every day, duplicated, has 6-nines up-time and in a hugely secure data centre), there’s a nervous twitch in my subconscious.

Anyway. I bit the bullet, ‘cos I’m rational, right…?

There’s a number of competing solutions, but I’d heard good things about Carbonite – very cheap ($50 a year for unlimited storage) , a 14 day free trial, encryption of all files, etc. So far it’s taken a few days for the initial backup to upload (there’s about 7Gb in the folders that I selected) but I haven’t noticed any performance or bandwidth hit. The interface is very intuitive (right click, choose whether to backup) – apart from that, Carbonite just sits there constantly monitoring your selected folders for any changes and trickling up any new or changed files. There’s another nice couple of features – first, if you accidentally delete a file, you can restore it for up to a month. Second, you can choose “back this up now” for particularly important files. One thing which would be incredibly useful which they don’t offer (but it may be in their development plan) is the ability to get to your files from another PC. This would be a bit like Box but with syncronisation. Maybe one day.

It’d be interesting to know whether this solution (or this type of solution) is being used by anyone for museum archiving, or even in a work scenario. Now that I’m starting to get my head around the fact that my files are out there, I’m liking it. In a nervous kind of way…