Backup Software and The Long Tail of Open Source

An addition to The Long Tail of “The Long Tail” related blogs :)

The Long Tail effect of open source has been discussed in several contexts. Some of the interpretations are likely very different from what Chris Anderson intended, but more and more niche needs of a relatively small set of people (which don’t have economies to justify product development by proprietary software companies) are being met by open source software. Witness more than 140,000 projects registered on SourceForge.

Whenever I install Red Hat Linux I am amused by some of the choices available for the language of the operating system - what is different when running a Red Hat box with English(Singapore) vs. English(India)!? It is also heartening to see support for regional languages making computers accessible to even remote villages (which don’t provide economic motivation to proprietary OS providers).

The long tail effect of open source has some interesting implications when it comes to backup software. Some of it was clearly visible in the recently completed survey of users of the open source backup software Amanda. The results are published on the Amanda wiki.

Let’s say your organization is using an operating system which is popular today but becomes out-of-favor in a few years’ time. It is very possible, actually very probable, that a proprietary backup vendor will withdraw support for this “unprofitable” operating system. This will force you to make a choice between either using some ad-hoc mechanism to backup that system or to replace the system with a different OS - both costly choices. Open source communities are known to provide support for older (and sometimes obscure) platforms. Furthermore, you have the source available to compile the software for your particular operating system. The Amanda survey found Amanda users using it to backup several operating systems which were either never supported by most proprietary backup software vendors, or whose support was dropped several years ago - e.g. DG UX.

In his article introducing the MySQL Storage Engine Architecture Arjen Lentz lays out the advantages of this architecture which enables a particular environment to choose how and where a database table is stored based on what is best suited for their particular situation. New storage and database innovations created to solve specific problems, require backup software to keep up. The long tail effect of open source software comes to the rescue here. For example, Zmanda Recovery Manager(ZRM) for MySQL enables a developer or a user of a storage engine to easily plug-in the best method for backing up their engine with ZRM.

Upcoming implementation of the Application API in Amanda will enable all application developers to integrate their applications seamlessly in a network backup environment. Truly open nature of this API (both from specification and licensing standpoint), enables its widespread use. In the past this integration between applications and network backup software has only been done by the vendors of backup software and hence been done only for widely used applications - ignoring The Long Tail of applications.

Similarly, the Device API of Amanda enables integration with new and innovative (or old and obscure) storage devices. We will be announcing one such very exciting integration in a few weeks. Stay tuned!!

————
Chander Kant

Comments are closed.