When Profits Hamper Innovation

I have been doing a lot of research in the area of Data Storage recently. The company I work for is a partner of Dell and we sell an awful lot of the Equallogic line of SAN/NAS devices. They have a great model, wherein a customer buys the disk array and is automatically entitled to use all the software related to the device !


This is also a very open model. Consider the fact that the devices contain
commodity hardware: SATA disks, SAS disks, and even Solid State disks, coupled with intelligent controllers that are easily upgraded. Add in firmware solutions including de-duplication, snapshots,copy on write protection, not to mention common protocols like iSCSI, CIFS, and NFS. Clients are left with a robust, resilient and scalable solution. They are priced very competitively in the market.

But that's not good enough for me ! So I started asking myself - is there an OpenSource solution ? The answer is "It depends...", with a gentle poke at my friend Bill Bitner at IBM. There are a couple of options, however they are slightly constrained.

The first one is called BTRFS (pronounced "butter"-FS). It is open source software, under the GPL copyleft license. It is undergoing rapid development, and it expected to be feature-rich in the very near future. Unfortunately, not all of the features are here yet. Which makes it an unviable solution for my corporate clients.

The second solution, pioneered by Sun Microsystems, is called ZFS. It has all of the features listed above, and more. Unfortunately, it was intentionally killed as an open source project when Oracle acquired Sun. Along with the operating system, called OpenSolaris, Oracle abandoned the open development of the filesystem.

ZFS is still available as an Enterprise solution, but in a closed-source, pay-to-play model. Oracle was able to do this, as the original work done by Sun was licensed under the much more restrictive CDDL. In essence, it allows the original authors to "own" any derivative works, and therefore close the source code at any time.

Oracle has in effect killed two well-received open source software projects, in favor of not competing with it's own interests. This is not news. Oracle has been watched with beat suspicion, as they acquired companies which had products that competed with theirs - Sun and BEA are just two examples. But in so doing, they hamper the innovation which is helping grow the limits of their own industry. What will happen to MySQL, or JAVA ?

Fortunately, a company called Nexenta has provided stewardship, and is actively maintaining the code, alongside a consortium of the original developers. The consortium effectively "forked" the open source components of OpenSolaris to a new project called IllumOS, and they continue to work with Nexenta on it's flagship, community project called NexentaStor.
NexentaStor turns an ordinary PC into a SAN/NAS device, very similar to the Equallogic device I described at the beginning of this post. It is feature-rich and open source, having a vibrant community continuing to provide new code and testing.




The opinions expressed in this post are purely those of the author. Opinions are like noses; everyone has one and they are entitled to it !