Jungle Disk Aids Amazon Storage

Amazon's Simple Storage Service, or S3, has emerged as popular storage option for Web-based businesses. However, S3 lacks a front end that's suitable for direct use by individuals and organizations looking to tap the service—and its attractive 15 cents per gigabyte per month pricing—for their own online storage and backup needs.

Enter Jungle Disk 2.0, a handy, $20 application that bridges the gap between Amazon S3 and your local desktop computer, and which works equally well on systems running Windows, Linux or OS X.

For its modest purchase price, Jungle Disk may be run on an unlimited number of separate computers, but all machines that run the software share a common account, and share common access rights to the data stored with Jungle Disk on Amazon's S3 service.

As a result, Jungle Disk is most appropriate for use by individuals or small groups that don't require multilevel permissions schemes. For organizations that require this sort of multiuser functionality, Jungle Tools is currently working on a Workgroup Edition of Jungle Disk priced at $2 per month per user. An initial beta version of Jungle Disk Workgroup Edition, which I have not yet tested, recently became available for download.

I've been using Jungle Disk for my personal remote backup needs for the past few months, both in its current 2.0 version, and in the version 1.5 incarnation that preceded it, and I've been impressed by the ease with which the application makes accessible Amazon's mostly developer-facing Web storage service.

Jungle Disk makes itself available on your machine in the form of a local WebDAV server that links back to a particular bucket (Amazon's term for storage partition within S3) that's configured for use with the software. You can connect to your Jungle Disk bucket through a local Web interface, through your operating system's built-in WebDAV client facilities—known as Web Folders on Windows—or by associating your Jungle Disk bucket with a mount point on your local machine.

With the addition of the $1 per month Jungle Disk Plus service, I was also able to access the files I'd uploaded to S3 from any location through a Web interface hosted on a server maintained by Jungle Disk, Inc. In addition to Web access for stored files, Jungle Disk Plus adds support for block-level file updates, in which S3 transfer costs can be minimized by allowing updates of only the changed portions of large files.

The Jungle Disk software maintains a local cache of your data, which speeds access of your files while the application works in the background to ferry data back and forth from S3. When you copy files to your Jungle Disk location, the software copies the files to its cache and then sets about syncing them to your bucket on the S3 service over an SSL-encrypted connection.

In addition to the SSL-secured link to S3, the files you transfer through Jungle Disk are themselves first secured with 256-bit AES (Advanced Encryption Standard)—using either your standard Amazon Web Services key or another key of your choosing.

The local cache size defaults to 1GB but is user-configurable. During my tests, I identified about 6GB of digital photographs that I wished to transfer to S3, and I set my cache size to 10GB to ensure speedy access to my files. With this arrangement, I was able to work with my data as if it resided locally while ensuring that my photos would survive a hard drive failure.

The one significant hitch I encountered with this setup was the absence of an offline mode in Jungle Disk that would allow me to work with files in my cache while disconnected from the service.

However, I could enable this sort of online/offline scenario by turning either to Jungle Disk's built-in backup functionality or by pointing a separate backup tool at my locally accessible Jungle Disk mount point.