It appears that this is in process … no, not simply the ZFS on FUSE, but a full fledged kernel subsystem.
This is interesting. ZFS is of course, the Sun file system which has had an altogether ridiculous amount of hype, while having a modest set of nice features. The Solaris and OpenSolaris was released under was not compatible with GPL, hence many people considered this OSino (Open Source in name only), as it was not legally possible to intermix the code between the largest GPL project (Linux) and the OpenSolaris code base. Likely anyone doing this would be on the wrong end of some legal action.
Which is why ZFS on FUSE was born. It provided an API that did not require the license admixture. FUSE generally works well … GlusterFS sits atop FUSE and works quite nicely (one of our favorite parallel file systems, and shipped by default on siCluster).
This development is important given the announced changes from Oracle, and the significant questions over the future of OpenSolaris. That is, if ZFS were confined to the Solaris platform, I don’t see it as increasing its market or mind share. I’d see that as a circling of wagons.
Of course, this appears to be rendered moot by the existence and development of BTRFS, which rethinks many things. ZFS is now owned by Oracle, and BTRFS is being developed by a large consortium of folks, including Chris Mason of Oracle.
If ZFS isn’t made into a compatible with Linux kernel system (FUSE or in-kernel based) with support from Oracle, then its likely not to have a long life when the other file system is stable enough for production use. Which, as of the 2.6.35 series, it might actually be. A few missing features, but not many.
If I were giving Oracle free advice (you know, worth its price), I’d suggest a license change on OpenSolaris, and a GPLing of it. This way, Solaris and OpenSolaris can take advantage of an influx of driver bits, of other development being done for Linux without pain. Because, otherwise, the relevance of OpenSolaris etc, ZFS inclusive, will continue to drop, rapidly. And this is definitely not in Oracle’s interest.