Replying to
@jaseg@chaos.social @coral@empty.cafe This those readout cards are stupidly expensive, next upgrade for HL-LHC they will be upgraded to next gen cards but yea even then.
Remote
EveyPN
@evey@chaos.social
mastodon 4.6.9Dutch, trying to learn german. And confused about everything. they/them. Shy. Feel free to interact or chat. Follow requests are okay,depending on profile and earlier interactions depends if i accept / C3NOC / 
425 Followers
362 Following
14 Posts
Joined October 14, 2024
Location:
Amsterdam
Matrix:
@evey:oneoftwelve.org
Dcache workshop time :3 #dcache
Replying to
@swope@mstdn.plus @Aaron_DeVries@mastodon.social the main issue is the Tdelta between active and power off temps. if both are equal you end up with lower retention. Documented in JEP122
Replying to
@jaseg@chaos.social @coral@empty.cafe See https://indico.belle2.org/event/10782/contributions/74815/attachments/27775/41084/LHCb_Giessen.pdf for some info on how they are used and where they are installed.
and https://indico.cern.ch/event/681247/contributions/2929079/attachments/1639220/2616679/PCIe40_Common_Readout_for_LHCb_and_Alice.pdf for more card specific info
Replying to
Replying to
@ave@social.treehouse.systems @ln@chaos.social @dx@mastodon.social @sa7dse@chaos.social and having standard filters that are easly sourced everywhere is quite nice too
Replying to
@Aurora_Bee@deadinsi.de Ye a good old bezuinigdmaatregel van de roverheid that is now beeing like helloooo
Replying to
@leyrer@23.social I mean there is practically no functional ECC on consumer platforms even if the CPU supports it the firmware is often lacking or just plain not functional or not using it.
Replying to
@spinach@girlcock.club Yee just not enough room for a big one, but i do know some companies in NL who will happly sell you a smol model :3
Replying to
@azrael@tech.lgbt @helle@tacobelllabs.net tbf Almere was build along the railroad with a reason.
Replying to
@job@bsd.network @swope@mstdn.plus @Aaron_DeVries@mastodon.social more like the energy difference of the electrons, the hotter they are the easier they migrate. The colder they are the slower they go. Or something like that
Replying to
@Codeberg@social.anoxinon.de We sometimes see similar numbers, or 10k+ when a user submits a 64core job in a single slot and the cgroup limiting kicks in. Bit annoying that load is a bit useless for that now a days
Replying to
@u880d@social.linuxkram.de factually incorrect tho CERN uses EOS for research data not CEPH.