[05:41] pav5088__ (~Mark@ppp118-208-18-239.lns20.bne1.internode.on.net) left irc: Quit: Leaving [06:50] simNIX (~simNIX@156-60.bbned.dsl.internl.net) left irc: Quit: Ik ga weg [07:01] ce^thue_thuet^ (~CE_KULIAH@200.199.88.88) joined #fai. [07:01] ce^thue_thuet^ (CE_KULIAH@200.199.88.88) left #fai. [07:12] pav5088 (~Mark@ppp118-208-18-239.lns20.bne1.internode.on.net) joined #fai. [07:58] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [08:05] justleen (~leen@rev-131-102.virtu.nl) joined #fai. [08:14] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) joined #fai. [08:55] i am thinking that we (at work) want a fai class SSD for solid state disks [08:56] they seem to need special attention for optimal performance and ware leveling. [08:56] has someone done that already? [09:08] Nic0 (~courtel@fw.tls.cena.fr) left irc: Quit: Leaving. [09:10] verwilst (~verwilst@router.begen1.office.netnoc.eu) joined #fai. [09:14] No. But I think it just a case of choosing the right file system. journaling file system seems to be a bad choice. [09:15] I would also add noatime to the mount options. [09:16] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Quit: leaving [09:17] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) joined #fai. [10:04] \sh (~shermann@nat.de.netviewer.com) joined #fai. [10:20] Nic0 (~courtel@fw.tls.cena.fr) joined #fai. [10:21] Nic0 (~courtel@fw.tls.cena.fr) left irc: [10:24] Nic0 (~courtel@fw.tls.cena.fr) joined #fai. [10:24] Nic0 (~courtel@fw.tls.cena.fr) left irc: [10:24] Nic0 (~courtel@fw.tls.cena.fr) joined #fai. [10:25] msc (~msc@gw1.cgn3.hosteurope.de) left irc: Quit: msc [10:37] msc (~msc@gw1.cgn3.hosteurope.de) joined #fai. [12:10] ErKa (~keryell@81.255.107.242) joined #fai. [12:43] Mrfai: aparently you also need to tell them that the operating system considers parts of the disk free [12:44] or rather, get the OS to tell them which parts are free [12:45] I think there are tools which can reset an SSD, so it knows everything is free. [12:45] Like a low level format [13:03] ah, i didnt know [13:03] but the problem is that a disk does not know that a file that was removed actually freed all the space that it once occupied. [13:04] the OS knows the space is free, but not the disk, by default. [13:05] and then i think batch writing (in long consequitive (?) chunks) is fast on normal disks and it is not on ssds. [13:07] msantana (msantana@SDF.ORG) joined #fai. [14:42] siggg (~jean@tra78-3-82-244-85-196.fbx.proxad.net) left irc: Ping timeout: 480 seconds [16:24] justleen (~leen@rev-131-102.virtu.nl) left irc: Quit: Leaving. [16:42] verwilst (~verwilst@router.begen1.office.netnoc.eu) left irc: Quit: Ex-Chat [16:54] ErKa (~keryell@81.255.107.242) left irc: Ping timeout: 480 seconds [17:54] Mrfai (~lange@malzmuehle.informatik.uni-koeln.de) left irc: Quit: leaving [18:00] pav5088_ (~Mark@ppp118-208-170-242.lns20.bne4.internode.on.net) joined #fai. [18:06] pav5088 (~Mark@ppp118-208-18-239.lns20.bne1.internode.on.net) left irc: Ping timeout: 480 seconds [18:56] ErKa (~keryell@pool-mlh-65.newel.net) joined #fai. [18:57] Mrfai (~lange@kueppers.informatik.uni-koeln.de) joined #fai. [19:05] pav5088__ (~Mark@ppp118-208-131-46.lns20.bne1.internode.on.net) joined #fai. [19:10] pav5088_ (~Mark@ppp118-208-170-242.lns20.bne4.internode.on.net) left irc: Ping timeout: 480 seconds [19:11] pav5088_ (~Mark@ppp118-208-181-168.lns20.bne4.internode.on.net) joined #fai. [19:15] pav5088__ (~Mark@ppp118-208-131-46.lns20.bne1.internode.on.net) left irc: Read error: Operation timed out [19:18] pav5088__ (~Mark@ppp118-208-177-141.lns20.bne4.internode.on.net) joined #fai. [19:25] pav5088_ (~Mark@ppp118-208-181-168.lns20.bne4.internode.on.net) left irc: Ping timeout: 480 seconds [20:05] fai: 3 lange committed revision 6196 to 7 fai: do not use fcopy for /etc/modules [20:48] Mrfai (~lange@kueppers.informatik.uni-koeln.de) left irc: Quit: leaving [21:38] stockholm: recent linux kernels support for TRIM [21:43] yes, i read about that, too [21:43] did you read about databases on SSDs in LWN the other week? [21:44] there was an article about an DB convention. OpenSQL or so [21:46] didn't read that, no [21:48] though ssd really rocks (l2arc, anyone? :)) [21:48] couldn't imagine kernel related work without ssd anymore ;) [23:42] \sh (~shermann@nat.de.netviewer.com) left irc: Remote host closed the connection [00:00] --- Wed Nov 24 2010