Building one big filesystem with JBOD and ZFS

Solution 1:

ZFS isn't designed to scale, or cluster, across more than one single Solaris (or other) system. It's not a distributed filesystem, so any distributed filesystem involving ZFS will have to have another layer that binds together the various mountpoints exposed by the individual systems in your cluster.

Why the ZFS insistance? Sounds like you want something more like the Hadoop filesystem.

Solution 2:

Add another server chassis to hold the drives. Pop drives and SAS expanders in it. For example the Chenbro RM91250 holds 50 drives; add two CK23601 expanders and a UEK for the expanders. Uplink that back to your main server (may need a SAS card with external ports if your server doesn't have one already). Or add a HP MDS600, might find one on eBay cheap-ish, and use it instead of the aforementioned box.

Note: I have no idea what you're using this for, so this might be totally inappropriate for your intended use. Fair warning this is just an idea, no guarantee of performance or suitability for a particular purpose.