Last year, say July time frame, if you asked me to name the top high performance computing file systems, and prognosticate who the up and comers were … well, you’d get lists much like I’ve said here in the past.
Lustre was the “king” and undisputed leader. pNFS was (sorry Bruce and team) effectively perpetually in the future (yeah, sort of like Perl6 … though we intend to play with both sometime soon … I hope). FhGFS was really niche. GlusterFS was an up and comer. Ceph was interesting, but way to early to see what it could do. There were some dark horses … Pomehlfs and others, that could make a case that they should have a seat at the table some time.
Now fast forward 8 months or so.
The Lustre community has multiple “leaders” none of whom want to say the magic F-word (fork, that is). Its future under Oracle has always been “fish or cut bait“. And, seemingly, they have cut bait. A valid business decision on their part, given the company, its direction, and its market set.
I am under NDA with a number of folks, so I can’t say much more about them. Lets just say that over the last month or so, I’ve come to the realization that some of the up and comers are not, at least in the HPC space.
Few of these things will alter our plans and efforts, though several of them are causing us to rethink some of our assumptions in the HPC storage space.
I do have to say that even though the poll is self selecting, and may not be a very accurate measure of reality, it is an approximation, and as one, could be interesting to interpret. It has caused me to rethink some things I thought I understood in the market. And that is a good thing.
Hopefully more soon. I’ll try to summarize what I can really say. In other cases, you’ll simply have to wait for the relevant group to make their announcements.