Commenti a: ZFS su Linux? certo che sì! http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/?utm_source=rss&utm_medium=rss&utm_campaign=zfs-su-linux-certo-che-si Quando ne saprò qualcosa, te lo farò sapere. Sat, 15 Oct 2011 14:08:48 +0000 hourly 1 http://wordpress.org/?v=3.3.1 Di: Chris Samuel http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/comment-page-1/#comment-13 Chris Samuel Fri, 27 Apr 2007 13:09:19 +0000 http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/#comment-13 Hi there, just checked back before heading off to bed (11pm here in Melbourne, Australia). :-) CDDL isn't compatible with the GPL (which may be why Sun created it) and so yes, it is unlikely that we will see ZFS in the kernel unless Sun has a change of heart and dual licenses it (which as the copyright holders they are perfectly at liberty to do). Still, ZFS under FUSE does some really nice stuff, including snapshots, and it's going to be really interesting to watch it develop. Chris Hi there, just checked back before heading off to bed (11pm here in Melbourne, Australia). :-)

CDDL isn’t compatible with the GPL (which may be why Sun created it) and so yes, it is unlikely that we will see ZFS in the kernel unless Sun has a change of heart and dual licenses it (which as the copyright holders they are perfectly at liberty to do).

Still, ZFS under FUSE does some really nice stuff, including snapshots, and it’s going to be really interesting to watch it develop.

Chris

]]> Di: EndelWar http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/comment-page-1/#comment-12 EndelWar Fri, 27 Apr 2007 10:43:20 +0000 http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/#comment-12 Thanks for your comment Chris, now I have the proof that trackback are useful :) In this post I comment a post published on another blog that talks about Open Source where they say that on the linux kernel there are licence limitations on included code, and that the CDDL code cannot enter the mainstream kernel, concluding the article saying that the only OS where ZFS can be used are FreeBSD and NetBSD, and on my post I link your article where you test the FUSE port demostrating that on linux ZFS can be tested and with good perfomance. Thank you for you test, I have to try ZFS on Fuse! Thanks for your comment Chris, now I have the proof that trackback are useful :)
In this post I comment a post published on another blog that talks about Open Source where they say that on the linux kernel there are licence limitations on included code, and that the CDDL code cannot enter the mainstream kernel, concluding the article saying that the only OS where ZFS can be used are FreeBSD and NetBSD, and on my post I link your article where you test the FUSE port demostrating that on linux ZFS can be tested and with good perfomance.
Thank you for you test, I have to try ZFS on Fuse!

]]> Di: Chris Samuel http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/comment-page-1/#comment-11 Chris Samuel Fri, 27 Apr 2007 10:04:10 +0000 http://endelwar.aregar.it/2007/04/27/zfs-su-linux-certo-che-si/#comment-11 Grazie! I don't speak Italian I'm afraid, but my wife translated your article as best she could. Whilst the patent issues do prevent use of ZFS under any license other than the CDDL ZFS/FUSE does work very well for me and I believe that Riccardo (who is doing the ZFS/FUSE port) is going to do a very good job of optimising it too. Some people believe FUSE to be a limitation for I/O performance, but I <a href="http://www.csamuel.org/2007/04/25/comparing-ntfs-3g-to-zfs-fuse-for-fuse-performance/" rel="nofollow">have some results</a> that show this is not necessarily the case. Grazie!

I don’t speak Italian I’m afraid, but my wife translated your article as best she could.

Whilst the patent issues do prevent use of ZFS under any license other than the CDDL ZFS/FUSE does work very well for me and I believe that Riccardo (who is doing the ZFS/FUSE port) is going to do a very good job of optimising it too.

Some people believe FUSE to be a limitation for I/O performance, but I have some results that show this is not necessarily the case.

]]>