I'm a big fan of Amazon Web Services (AWS). I use Amazon S3 to backup my websites every day and have used the CloudFront as an inexpensive CDN for image hosting. Lots of individuals and businesses also use S3 as a cloud-based backup too, using Jungle Disk, S3Fox, and other utilities to transfer files directly to S3.

But what happens if you have A LOT of data to transfer. Like hundreds of gigabytes, or multiple terabytes? If you want to take advantage of the various AWS services -- say you want to backup all of your home movies for double-security/potential global access or your business has a few terrabytes of data it would like to incoporate with an EC2 application -- how long is it going to take to actually transfer that much data? It almost seems like it would be faster to just ship Amazon a hard drive and let them transfer everything. Which is exacty what Amazon is proposing you do with its new AWS Import/Export service.

The service, which is in limited beta (you can apply now) and currently only uploads to US-based S3 buckets for now, works like this:

• Fill out an electronic form detailing your S3 bucket info, AWS password and a shipping address
• Attach some signed digital files to your external device
• Mail your external hard drive, with all required power adaptors and connection cables to Amazon.

Once the drive is received, the data will be transferred to your AWS account (the type of drive you send can impact the data transfer speed, for instance, eSATA will be faster than USB) and the drive will be mailed back to you.

The price? $80 per device, plus $2.50 for each hour of transfer required. Right now, the service only supports sending data to Amazon, but in the future, Amazon will support sending external drives with large amounts of data back to customers, similar to what Backblaze offers its customers as a recovery solution.

Amazon has a great pricing calculator so you can work out how expensive a large transfer might be as compared to standard S3 upload bandwidth costs (which are not charged for AWS Import/Export transfers) and the detaild description section of the AWS Import/Export page lists the transfer time for uploading 1 TB of data at various connection rates.

If you have lots of data you would like to host on Amazon's secure cloud, AWS Import/Export might just be the right tool for the job!

Amazon has just launched Elastic Compute Cloud (or EC2), a "Tier 0" service in the same vein as S3 (Simple Storage Service, launched earlier this year), but for processing power instead of storage. Amazon describes EC2 as "a web service that provides resizable compute capacity in the cloud ... designed to make web-scale computing easier for developers." Developers can run anything they want on EC2--web servers, database servers, game servers, 3D renderers--though currently only Linux is supported. The way it works is this: You create an Amazon Machine Image (AMI) which contains your environment, programs, data, and so on and upload it to EC2 (or choose one of Amazon's "pre-configured, templated images), then use the EC2 web service to configure its security and network access. Then you can programatically "start, terminate, and monitor as many instances of your AMI as needed" on the fly. According to Amazon, each instance "predictably provides the equivalent of a system with a 1.7Ghz Xeon CPU, 1.75GB of RAM, 160GB of local disk, and 250Mb/s of network bandwidth," which is, well, a whole lot of power.

For what you get, EC2 is fairly inexpensive, though that's to be expected having seen the pricing for S3. Here's the rundown:

• Pay only for what you use.
• $0.10 per instance-hour consumed (or part of an hour consumed).
• $0.20 per GB of data transferred outside of Amazon (i.e., Internet traffic).
• $0.15 per GB-Month of Amazon S3 storage used for your images (charged by Amazon S3).

If you're using it 24/7, that comes out to about $74 per month per instance, which isn't bad when compared to a dedicated server with similar specs. And if you're not using it constantly, the deal is even sweeter. The biggest advantage, though, is that you can start and stop new instances on the fly: Whereas setting up an additional dedicated server can take days, with EC2 it takes minutes, and since it can be controlled programmatically, you can set you programs up to automatically start or stop instances without your interaction.

Amazon S3, the online storage service from the mega-retailer, has been on my radar for a couple of months, but unless I'm mistaken, JungleDisk is the first desktop-based (rather than web-based) app that takes advantage of the service. Basically, JungleDisk integrates S3 storage with Windows, Mac OS X, or Linux allowing you to manage files stored on the service as though they were ordinary files on a regular hard drive. JungleDisk's intended use is for backing up important files and transferring files between computers, and the web site has a handy calculator for figuring out the cost depending on your use. Want 20GB of storage and 2GB transfer? $3.40/mo. Want 500MB of storage and 2GB transfer? $0.48/mo. Apart from Amazon's fees, JungleDisk itself is free.