The open source Lustre high performance computing file system isn’t on the list of forked projects, but it is getting a new, community-driven distro that is trying really hard to say that they’re not officially a fork. ‘Since April of 2010 there has been confusion in the community, and we’ve seen an impact in the business confidence in Lustre,’ Brent Gorda, CEO and president of Whamcloud told InternetNews.com. ‘The community has been asking for leadership, the commitment of a for-profit entity that they can rely on for support and a path forward for the technology.'”
Paraphrasing Shakespeare, a fork by any other name …
Look … I appreciate that no one wants to call this a fork. Oracle has seemingly abandoned the project and is shopping ownership of the IP around. The choices ahead of the community are find someone to buy the IP, and rally to their leadership, ignore the IP, rename the project and fork it. You could always pretend that the IP isn’t an issue, that no fork is needed, and then have to do some serious rhetorical contortions to explain why your release isn’t a fork.
And that gets to the second aspect of this. Linux as a project is successful as there is a single strong guiding hand, a community of functionality developers and system maintainers, and a well defined heirarchy/mechanism to interact with the community. We all benefit from this organization.
Lets be honest. It doesn’t (yet) exist within the Lustre community. I’d like it to exist, but people are getting pissed at me for noting the obvious … the 800 pound gorilla in the room. Oracle owns the IP. If you release code without their buy in, are we gonna call it a spork?
As I noted in the previous post, this structure needs to form. I’d like to see one group own it, take leadership, and then people follow on from there. This is the best way to guarantee that there is a single future and not a fractured future.
It is worth reading this LWN article carefully. Look at what happened with these other projects. Could it happen here?
That was the point of the previous post. It could happen, without strong leadership.
I might suggest that the best course of action would be for Whamcloud to negotiate the IP ownership with Oracle, giving Oracle a fraction of the company for the IP, as an investment. This would be the best course of action, as it would enable a coalescing around one group.
Right now, we have 3 different organizations, and 2 different for profit companies, all working on Lustre. And none of them “owns” it. Thats a 3rd party, who isn’t seemingly interested in continuing to do so. If this situation persists, and there is no change in IP ownership, and 3 different groups, then we could have problems.
Everyone will cooperate, until a Whamcloud customer wants X, while a Xyratec customer wants Y, with the two being not compatible. Which is my point.
And of course, while this is all playing out, viable competitors are on the horizon and available now. GlusterFS 2.x wasn’t great. 3.x is pretty good, and has features Lustre doesn’t. Ceph, which is in the kernel, is already spawning other very useful projects (RBD and others), and has a very nice feature set relative to Lustre. There is FhGFS, which does distributed metadata, but isn’t open source. There are others looming out there as well.
Since we sell systems pre-installed with Lustre as well as other cluster file systems, and we have customers (current and prospective) with a business dependency upon Lustre, seeing this resolved properly is in our interest.
Lustre has enough challenges ahead of it, challengers today and looming. It has a future, but I am hoping its not a bifurcating or worse future. Lets call the fork a fork, call it WFS (Whamcloud FS) or something, and go from there.