Data storage and backup are critical to any organization's IT infrastructure. Just having a large, external drive attached to your office server isn't enough. While drive reliability has gone up, and costs have gone down, local (in your office) external storage is still subject to the weird and unexpected. Disks still crash, data is still corrupted and systems can be hacked. Worse, your office could be flooded (recall Tropical Storm Sandy) or you could be hit with a fire in your building. With more and more of your business being run on software, you need to make absolutely sure you have multiple, redundant backup stored at different locations. We not only backup all our web servers and data servers fully at our data center, but we also backup EVERYTHING to the cloud with Amazon S3.
There are great backup services out there like Mozy and Carbonite. They are easy, fairly inexpensive and reliable. But for this post, I am focussing on Amazon S3.
As mentioned in the last post, Amazon offers high-end IT services to the masses. S3 is one of their early services offered as a way developers could store and retrieve data. Then, as time went on and the service matured, it became a great way to backup, archive and recover. You can even use it to host static web pages with 99.99% uptime.
Amazon boasts "99.999999999% durability, with 99.99% availability" which is as good as it gets. Their design requirements are "Secure, Reliable, Scalable, Fast, Inexpensive and Simple" and they live up to that requirement.
Pricing is broken into three pieces: Storage, Requests and Transfer. Storage means how much data you put on their hardware. Storage costs $0.03 per month per gigabyte. Thats insanely cheap. That means 1 terabyte (1000 gigabytes) of storage would cost you $30 per month. What's better is that they only charge you for the amount you used in that given month. Explaining Requests and Transfer costs gets a bit geeky, but suffice to say its very inexpensive as well.
You need an Amazon account to use S3 and management of the storage is done through their web-based console. Storage is done in a "bucket" model where you create a bucket for any specific storage requirement you have. There are great third party "complete' back-up packages to automate periodic backups and recovery if it is needed. We use a product call Cloudberry.
If you're looking for a very safe and reliable cloud storage provider, think Amazon S3. We always recommend hiring an IT professional to set up and manage your infrastructure. But knowing about S3 and suggesting its use to your IT company will save you money. More importantly, using S3 will let you sleep at night knowing that the Amazon Tier IV, top-of-class infrastructure, is taking care of your data.
