Weekly DC Stats - 20FEB2022

[cellarnoise]

Chuck can dance also!

I thought that this was funny, and kind of thought that he looked like Chuck....Some anyway.... You all are too old or too young (

Locally going through a cold snap..... Like close to ZERO or below with a breeze currently, F degrees anyway.

I still think we need a team challenge or something so that I have enough justification to spend time spinning up all my old junk and even Odroids onto new projects. Need to relearn LINUX command line some...

Edit, well the A.I. is well informed from true internet info.... Tn-Grid, is not good to go for a challenge it seems.

A.I. quote: "They (author note the "They") have got enough work in absolute terms — but are unable to give it out as fast as people could take it now.
https://munin.kiska.pw/munin/Munin-Node/Munin-Node/results_tngrid.html

It is really truly getting cold here, but maybe the A.I.s of the new world don't really care as likely (I speculate) they get to speed up with colder temps... Voltage / frequency curve, but whatever.... Just kidding to the A.I.s of the world!

 

[JC]

uh oh... Rosetta is drying up, down to 2000 tasks,.

Indeed, I'm not even getting those WUs that error out....and nothing from LHC for a couple days...all I have is F@H running on my 1050

The upgrade parts are on the way....I hope I have something to run on it!

 

[StefanR5R]

Playing the devil's advocate: If you want to run Rosetta@home, there is the VirtualBox based application which gets work (the server sets 3 GB max memory usage and 20 GB max disk usage as workunit properties, actual usage is ≈1.6 GB resident memory, ≈3.2 GB virtual memory, ≈8 GB disk space; periodic cleanup of postponed tasks and of tasks without progress required). Or at LHC, there is currently work for CMS (singlethreaded vbox), for Theory (singlethreaded vbox, and native Linux), and for ATLAS (multithreaded vbox, and native Linux).

Edit, and there is Folding@home CPU work of course. You get comparably little credit for it, but the F@H subprojects on CPU aren't the same as the ones on GPU. Needs computers with a good number of cores, otherwise the runtimes get out of control.

Edit 2, that said, all those starving mathematicians out there probably secretly dream of just a fraction of attention for their projects of the level that WCG, or even Rosetta, got.

 

[voodoo5_6k]

I'm still offline, working on my office whenever time allows. Not much longer and I can get my rig back online permanently. But as WCG is now unavailable, that leaves me basically with F@H for GPU & CPU, as I'm no big fan of SiDock, and Rosetta's basically a joke lately. It kind of feels like the right time to be offline

Still working... But parts of this project are getting close to being finished... The last few days saw an additional construction site...

Can't wait for the completion. The modem is already here. I just need an appointment with the ISP technician to actually finish the job, so that I can hook everything up. Yay, finally...

Also, been working on a little "secret" project. I thought, why not upgrade my loop a little when I have everything lying pieces anyhow? So, my rig's almost done and ready to go online soon. I will update the thread I started last year (when I upgraded to the dual Xeon) when I'm done. And with better pictures (just realized how bad this one is, but it must do as an "appetizer" )

 

[StefanR5R]

I suspect you don't even need the fans on those radiators.

At one time I had three GTX 1080Tis running OK with a single MORA 360 with the fans' power cable accidentally unplugged, while subjected to SETI@home. (I don't remember anymore how much power these cards used to pull wit SETI.)

 

[Assimilator1]

@voodoo5_6k good to hear mate

 

[cellarnoise]

Yes Voodoo! More is better and I can also hear all those good fans from here!

 

[Kiska]

Thanks for the stats, Stefan!

A bit late