**** BEGIN LOGGING AT Tue Jan 25 02:59:57 2011 Jan 25 08:08:35 Morning. Jan 25 08:59:37 hey mvzaanen Jan 25 08:59:44 hey Serdar Jan 25 08:59:56 going to fosdem again? Jan 25 09:00:20 yes, but we don't have space for any devrooms :/ Jan 25 09:00:33 we don't have space? Jan 25 09:00:38 there is no space? Jan 25 09:00:42 or did we not apply for space? Jan 25 09:00:50 I did Jan 25 09:00:56 I asked for space Jan 25 09:01:01 ah ok Jan 25 09:01:04 but we were declined Jan 25 09:01:12 it was on oktober Jan 25 09:01:44 ah, I see Jan 25 09:01:45 pity Jan 25 09:03:08 yes Jan 25 09:03:13 are you going there? Jan 25 09:03:33 I don't think so Jan 25 09:03:39 colleagues here were asking me as well Jan 25 09:03:46 weekends are typically family time Jan 25 09:03:56 and on saturday an old friend of mine is coming as well Jan 25 09:05:24 I am going to find some guys there to ask about b-trees and filesystems Jan 25 09:05:53 b-trees? Jan 25 09:05:59 balanced? Jan 25 09:06:01 I need to know, why b-trees are well-designed for filesystems Jan 25 09:06:04 bayer Jan 25 09:06:05 I vaguely remember that stuff Jan 25 09:06:08 oh... Jan 25 09:06:13 but bayer tees are balanced too Jan 25 09:06:18 trees Jan 25 09:06:24 I remembered learning about w-trees as well Jan 25 09:06:28 which are also balanced Jan 25 09:06:32 and easier to keep balanced Jan 25 09:06:33 w-trees? Jan 25 09:06:36 that was years ago Jan 25 09:06:42 I don't think they really picked up Jan 25 09:06:53 b-trees are more well knowned I guess Jan 25 09:06:59 yeah Jan 25 09:07:16 I think something like b+ or b* trees exist as well Jan 25 09:07:23 reiser, btrfs and many other filesystems are yousing b- b+- and b*- trees Jan 25 09:07:26 it's been years since I read about that Jan 25 09:08:44 me too, but I need it now again Jan 25 09:09:36 why? what are you doing with that? Jan 25 09:09:59 filesystem Jan 25 09:10:28 or better, I wrote a distributed in-memory filesystem Jan 25 09:10:37 ah, cool Jan 25 09:10:38 and now I need to make it better Jan 25 09:10:47 I am working on the metadata Jan 25 09:11:23 which I want to put on a b-tree, because the access time of a b-tree is phenomenal Jan 25 09:12:25 how much meta-data goes in there? Jan 25 09:12:33 do you really need to have it balanced? Jan 25 09:13:05 I was thinking about it Jan 25 09:13:36 I want put there many many files Jan 25 09:14:06 this is why I need to store many many inodes Jan 25 09:15:25 how can I write many inodes and read them very fast? Jan 25 09:15:33 that's my priority Jan 25 09:15:54 what I have (now) is a kind of binary tree Jan 25 09:16:00 root node / Jan 25 09:16:19 which only has one first_child Jan 25 09:16:40 and ich child has siblings (same directory) Jan 25 09:16:49 s/ich/each Jan 25 09:17:02 and each child has first_child Jan 25 09:17:37 if first_child exist == directory Jan 25 09:17:45 if not == regular file Jan 25 09:18:09 this is the actual situation Jan 25 09:26:50 ah, I see Jan 25 09:27:15 yes, you'd like to have that balanced Jan 25 09:32:39 yes, but I would prefer not balancing ... Jan 25 09:33:04 because distributed balancing... Jan 25 09:33:13 I don't know Jan 25 09:33:42 I would love a balanced tree without this balancing stuff :D Jan 25 09:37:27 balancing is annoying anyway Jan 25 09:42:37 I know Jan 25 21:15:33 anyone know if Serdar is around? Jan 25 21:22:02 Aitch_, 12:12* Serdar (~Serdar@ruiz.cs.uni-duesseldorf.de) has joined #android-on-freerunner Jan 25 21:22:02 15:23* Serdar has quit (Quit: This computer has gone to sleep) Jan 25 21:39:00 Thanks Flack, do you know the latest re daily images? **** ENDING LOGGING AT Wed Jan 26 02:59:57 2011