> just have 2 computers that are going to be using the data,
It will first be two, but then should grow up to "several" nodes.
> try just a cluster{ed|ing} file system.
That's what I'm after. How is CODA not a "cluster{ed|ing}"
filesystem?
> OpenGFS,
http://opengfs.org has been down (for the last few days at least),
there has been only 1 message posted to the support mailing list this year
(in April!), and no updates released since January. Does not seem
suitable because there is not a real-world production user base that I can
find.
> Oracle Cluster File System,
Too expensive for this project.
> or one of the others.
What others? I'm willing to consider alternatives to CODA, but I
know of none. OpenAFS's replication is read-only, not read-write, and
frankly seems as admin-heavy as CODA. Also, this cluster is an experiment
in complete node homogeneity. I want to use the big hard drives of my
nodes instead of additional hard drives on yet-more-hardware ($$$).
--Derek
