Question

Posted Oct 13, 2008 19:55 UTC (Mon) by kragil (guest, #34373)
In reply to: Question by corbet
Parent article: Linux Summit will preview new advanced file system (SearchEnterpriseLinux.com)

Ok I see, thank you.
It is all available though ( http://tux3.org/ ).
But it is bitbucket/Mercurial hosted. That might explain a lot ;)

Question

Posted Oct 13, 2008 20:28 UTC (Mon) by rahulsundaram (subscriber, #21946) [Link] (7 responses)

Usually, hosting it in a external repository somewhere just doesn't cut it. It needs to be posted to LKML as a series of patches allowing specific feedback and then respond with more iterations till it gets into a acceptable state.

Question

Posted Oct 13, 2008 21:57 UTC (Mon) by njs (subscriber, #40338) [Link] (6 responses)

btrfs hasn't been posted for inclusion either, though, has it?

Question

Posted Oct 13, 2008 21:58 UTC (Mon) by rahulsundaram (subscriber, #21946) [Link] (5 responses)

It has been.

http://thread.gmane.org/gmane.linux.file-systems/26246/fo...

Question

Posted Oct 13, 2008 23:04 UTC (Mon) by kragil (guest, #34373) [Link] (4 responses)

Interesting thread (especially Teds mail). So Btrfs is pushed by a lot of companies... And the masterplan is to go from ext3 to ext4 and then to btrfs.
OK, but I still think it is good to have a choice. Having a good selection of feature rich filesystems will not hurt. Distros pick their favorites anyways.

Question

Posted Oct 14, 2008 4:16 UTC (Tue) by jamesh (guest, #1159) [Link] (2 responses)

Note that most distributions have settled down on ext3 at the moment. I don't see many trying to differentiate themselves on the basis of the default file system choice.

That might change post-ext4, but then again distributions might continue to cooperate. If we assume that most distributions switch to ext4 in the short term (this seems likely due to the low cost of switching), that leaves a decent amount of time for both btrfs and tux3 to try and prove themselves.

Question

Posted Oct 14, 2008 17:46 UTC (Tue) by drag (guest, #31333) [Link] (1 responses)

Ext3 vs Ext4 is going to be a hard transition for distributions because it's not obvious that Ext4 is going to have lots of improvements that will have a substantial impact on end user experience. Does 'now have extents' really going to make Openoffice start faster or make my system more 'crash resistant'?

Btrfs, on the other hand, has that really 'wow neat features' aspect to it. The same sort of attractive stuff that makes people lust after ZFS without ever actually touching it or knowing that much about it or whether or not it will eat their data or run with any sort of acceptable performance.

It seems pretty likely that, at this point, people would end up supporting Ext4 for the minority of people that specifically request it and leapfrog it by making Btrfs the real successor to Ext3 (by way of the default FS selected by their installers).

Question

Posted Oct 15, 2008 0:13 UTC (Wed) by qg6te2 (guest, #52587) [Link]

Significantly shorter fsck times for Ext4 is a compelling reason by itself. Reduced fragmentation and being able to handle more files as well as larger hard drives is yet more reasons. As good as btrfs is going to be, it is years away from completion ("production ready" is being touted for 2012), vs Ext4 that is just around the corner.

Ted's email

Posted Oct 16, 2008 13:32 UTC (Thu) by bluss (guest, #47454) [Link]

Ted's email on ext4/btrfs is indeed very interesting, even carrying old but not previously known news. Thank you!

http://thread.gmane.org/gmane.linux.file-systems/26246/focus=26317

Snippet

As far as btrfs is concerned, one of the things that you may not know
is that about a year ago (on November 12-13, 2007), a small group key
filesystem developers, that included engineers employed by HP, Oracle,
IBM, Intel, HP, and Red Hat, and whose experience included working
with a large number of filesystems: ext2, ext4, ext4, ocfs2, lustre,
btrfs, advfs, reiserfs, and xfs came together for a two day "next
generation filesystem" (NGFS) workshop.  At the end of the that
workshop, there was unaminous agreement (including from yours truly)
that (a) Linux needed a next generation filesystem to be competitive,
(b) Chris Mason's btrfs (with some changes/enhancements discussed
during the workshop) was the best long-term solution for NGFS, and (c)
because creating a new enterprise filesystem always takes longer than
people expect, and even then, it takes a while for enterprise users to
trust a new filesystem for their most critical data, ext4 in the next
generation of filesystems was needed as the bridge to the NGFS.