| |||||||||||||
|
| |||||||||||||
Gee that's tough.Gee that's tough.Posted Mar 20, 2007 11:58 UTC (Tue) by nix (subscriber, #2304)In reply to: Gee that's tough. by qu1j0t3 Parent article: The 2007 Linux Storage and File Systems Workshop
They're ignoring the ideas of ZFS. Right.
Of course this means that the Val Henson who's a Linux filesystems hacker must be an *entirely different person* from the vhenson@eng.sun.com who was closely involved with ZFS development? (Perhaps she has a secret twin.)
(Log in to post comments)
Gee that's tough. Posted Mar 20, 2007 18:10 UTC (Tue) by zooko (subscriber, #2589) [Link] The fact that Val Henson worked on ZFS doesn't prove that the Linux filesystem developers are learning lessons from ZFS. It doesn't even prove that Henson has learned lessons from ZFS. Indeed, since Henson is the author of the "Compare-By-Hash" paper critiquing compare-by-hash on intuitively appealing but incorrect grounds, perhaps she dislikes the ZFS design, which features compare-by-hash among many other ideas. Note: whenever anyone cites that regrettable compare-by-hash paper, they really ought to cite Graydon Hoare's follow-up (disclaimer: I helped Graydon a bit on writing that page), John Black's follow-up, and Henson and Henderson's much better self-followup. Oh, I see that Jeff Bonwick is given thanks in that last paper. This is evidence that Henson has benefitted from the lessons of ZFS.
Gee that's tough. Posted Mar 31, 2007 0:57 UTC (Sat) by wmf (guest, #33791) [Link] But ZFS doesn't use compare-by-hash; that's one of its advantages over previous work like Venti/Fossil.
|
|
||||||||||||