F@H Team "hardware.no" (lagnr: 37651)

[tore_v1]

Jeg kan sende deg en cd.. Send adresse på pm..

 

Dette her er lagtråden til hw.no, så da må du nesten forholde deg til det. Hadde vært flott å få enda en q6600 med på laget!

 

Selv er hk'et mitt kaputt.. Sender det inn til retur på mandag.. Håper det går fort unna!

[Tordenflesk]

Har ikkje fleire tomme cd'r...

Har du blanke DVD-er kan du bruke Imgburn og brenne imaget til en DVD.

[jon67]

Har ikkje fleire tomme cd'r... og ville være dumt å waste på starta wu'r bare for å skifte lag! uakseptabelt!

9551748[/snapback]

Helt enig. Heldigvis går en SMP-WU unna på 12-15 timer på en overklokket Q6600, og da...?

[rvenes]

Njaa... nå har eg spelt litt i dag, men alikevel skulle den da fått 2 og ein halv cpu i allefall... Kva med "bigpackets" eller er det ikkje noe som heiter det på smp saken?

 

Ser at det er mykje kortere tidsfrist på smp versjonene enn på dei vanlige pakkene.

 

SMP:

Download time: September 22 11:07:14

Due time: September 26 11:07:14

 

Vanlig:

Download time: September 15 00:53:50

Due time: December 20 00:53:50

 

Vis ein får meir poeng for SMP versjonen, så går det vel fint å køyre den i stade for 2 vanlige, så lenge den blir ferdig innen tidsfristen? Vil tru at ein E6600 ville klart å fått ferdig ein SMP sak i løpet av 4 dager.

 

Det er servern min det er snakk om, den har fine dager, så lenge den ikkje driver med folding,

Endret 22. september 2007 av Cornel

[jon67]

Alle smp-wuer er big packets, men noen er større enn andre

[madammim]

Hei.

 

Har mistet 3 WU på rad på min SMP. 2 av de var over 90%

 

På den ene fikk jeg opp melding om at det hadde gått lang tid uten fremgang... Eller noe slik. På de 2 andre sjekket jeg og fikk bekreftet at prosessorforbruket var 0. Restartet maskinen og fikk feilmelding når klienten startet.

 

Jeg har alltid bare restartet maskina, og det har gått bra. Men etter disse skal jeg slutte med det.

 

Er "ctrl + v" sikker når man slukker klienten?

 

EDIT: På min E6400@3400 med 3GB RAM får jeg ca 1957 ppd (fahmon). En WU er unnagjort på vel 20 timer.

 

EDIT: Mente "ctrl + c", ja

Endret 29. september 2007 av madammim

[Fruktkake]

CTRL + C er lukk.

[Gaervern]

Kjørte en gang -config på smp-klienten når den hadde hengt seg opp. Da kom den i gang igjen. Jeg garanterer ingenting, men omstart av maskinen er uansett risikabelt.

[Jimmarn]

har dete på ps3 jeg

[O.J]

PS3 folding er jo alltids bra.

[tore_v1]

Godt du er med Detrix, fyr opp foldingen på PC-en også, så skal du se at du klatrer oppover lista

[Gaervern]

PS3ene nådde nettopp en samlet regnekraft på 1 petaflop, like mye som de kraftigste superkomputere som er laget.

 

http://fah-web.stanford.edu/cgi-bin/main.py?qtype=osstats

 

Samlet har FAH 1,2 petaflop, for en uke siden var det bare 0,9. Oppdateringen av PS3 klienten og de nye WUene som utnytter flere kjerner gir en enorm økning i regnekraft og større fleksibilitet. Neste gjennombrudd kan komme når en GPU-klient som like brukervennlig som PS3-klienten blir ferdig.

[O.J]

Med nyvinninger innen GPGPU funksjonalitet kan nok det bli en realitet.

 

Men jeg er også skeptisk til at PS3 og GPU'er bare kan kjøre et begrenset antall WU'er.

Noen som vet hvor stort problem det er for prosjektet?

[Gaervern]

Det virker som om de er entusiastisk over de vitenskapelige mulighetene økningen i regnekraft betyr (FAH nærmer seg nå 1,3 petaflops):

 

"New PS3 version results looking very nice scientifically

 

It's still early, but the newest PS3 version (1.3) is looking very nice scientifically. While the key science upgrades in 1.3 are very much behind the scenes, they are going to play a big role in our work. In particular, the main code to describe how the protein interacts with water was majorly upgraded (to Generalized Born or GB) and we're getting very nice results with it, but still at a great speed up over a PC.

 

One interesting side effect of this new GB code is that since the code is more sophisticated scientifically, it uses the Cell processor's ability to do lots of FLOPS much better. One major challenge for PS3 and GPU programming is just getting the data to the CPU. Once that's done, in many cases (for us), the flops are almost free (or at least very, very efficient).

 

So, the upshot is that the new PS3 client in GB does a lot more flops per client, often as many as 35GF in some calculations running in the lab. As we roll out these new GB calcs more and more to FAH/PS3, the performance will go up, and the science we're getting back will be much more useful, so it will be exciting for us in lots of ways."

http://folding.typepad.com/news/2007/09/new-ps3-version.html

 

Edit: antall aktive PS3er har gått opp fra 33 500 for noen dager siden til over 41 000, antakelig som følge nyhetene om framgang, og det har resultert i at de alene leverer 1,030 petaflop!

Endret 24. september 2007 av Gaervern

[Xell]

Hei.

 

Har mistet 3 WU på rad på min SMP. 2 av de var over 90% 

 

På den ene fikk jeg opp melding om at det hadde gått lang tid uten fremgang... Eller noe slik. På de 2 andre sjekket jeg og fikk bekreftet at prosessorforbruket var 0. Restartet maskinen og fikk feilmelding når klienten startet.

 

Jeg har alltid bare restartet maskina, og det har gått bra. Men etter disse skal jeg slutte med det.

 

Er "ctrl + v" sikker når man slukker klienten?

 

EDIT: På min E6400@3400 med 3GB RAM får jeg ca 1957 ppd (fahmon). En WU er unnagjort på vel 20 timer. 

9553754[/snapback]

 

du kan prøve denne fremgangsmåten. Vet den står under Linux, men jeg bruker qd-tools under windows også. Har ikke prøvd queuefix for smp under windows, men det kan jo være verdt ett forsøk.

[Gaervern]

Det er opplegg for flere studier enn det er prosessorer, selv om man har planer om å få atskillig mer ut av PS3er og GPUer i framtida, også når det gjelder fleksibilitet. Fold on!

 

http://forum.folding-community.org/fpost199353.html#199353

[Gaervern]

Fruktkake, du bør poste loggen din med de kjapt ekspederte WUene her:

 

http://forum.folding-community.org/fpost199355.html#199355

[rvenes]

Hmm... hadde ein WU som stoppa på 37% så sendte den inn resultatet, og downa ein ny og forsatte?

 

Logg:

 

Klikk for å se/fjerne innholdet nedenfor

[05:50:41]

[05:50:41] *------------------------------*

[05:50:41] Folding@Home Gromacs Core

[05:50:41] Version 1.90 (March 8, 2006)

[05:50:41]

[05:50:41] Preparing to commence simulation

[05:50:41] - Assembly optimizations manually forced on.

[05:50:41] - Not checking prior termination.

[05:50:42] - Expanded 397842 -> 2040649 (decompressed 512.9 percent)

[05:50:42] - Starting from initial work packet

[05:50:42]

[05:50:42] Project: 3402 (Run 6, Clone 756, Gen 28)

[05:50:42]

[05:50:42] Assembly optimizations on if available.

[05:50:42] Entering M.D.

[05:50:48] Protein: p3402_dpdp-ii TIP3P ff94

[05:50:48]

[05:50:48] Writing local files

[05:50:48] Extra SSE boost OK.

[05:50:48] Writing local files

[05:50:48] Completed 0 out of 4000000 steps (0%)

[06:05:48] Timered checkpoint triggered.

[06:20:49] Timered checkpoint triggered.

[06:23:19] Writing local files

[06:23:19] Completed 40000 out of 4000000 steps (1%)

[06:38:20] Timered checkpoint triggered.

[06:53:19] Timered checkpoint triggered.

[06:55:50] Writing local files

[06:55:50] Completed 80000 out of 4000000 steps (2%)

[07:10:50] Timered checkpoint triggered.

[07:25:51] Timered checkpoint triggered.

[07:28:19] Writing local files

[07:28:19] Completed 120000 out of 4000000 steps (3%)

[07:43:19] Timered checkpoint triggered.

[07:58:19] Timered checkpoint triggered.

[08:00:49] Writing local files

[08:00:49] Completed 160000 out of 4000000 steps (4%)

[08:15:49] Timered checkpoint triggered.

[08:30:49] Timered checkpoint triggered.

[08:33:18] Writing local files

[08:33:18] Completed 200000 out of 4000000 steps (5%)

[08:48:18] Timered checkpoint triggered.

[09:03:18] Timered checkpoint triggered.

[09:05:48] Writing local files

[09:05:48] Completed 240000 out of 4000000 steps (6%)

[09:20:48] Timered checkpoint triggered.

[09:35:48] Timered checkpoint triggered.

[09:38:17] Writing local files

[09:38:17] Completed 280000 out of 4000000 steps (7%)

[09:53:17] Timered checkpoint triggered.

[10:08:17] Timered checkpoint triggered.

[10:10:47] Writing local files

[10:10:47] Completed 320000 out of 4000000 steps (8%)

[10:25:47] Timered checkpoint triggered.

[10:40:48] Timered checkpoint triggered.

[10:43:17] Writing local files

[10:43:17] Completed 360000 out of 4000000 steps (9%)

[10:58:18] Timered checkpoint triggered.

[11:13:19] Timered checkpoint triggered.

[11:15:48] Writing local files

[11:15:48] Completed 400000 out of 4000000 steps (10%)

[11:30:49] Timered checkpoint triggered.

[11:37:29] - Autosending finished units...

[11:37:29] Trying to send all finished work units

[11:37:29] + No unsent completed units remaining.

[11:37:29] - Autosend completed

[11:45:49] Timered checkpoint triggered.

[11:48:18] Writing local files

[11:48:18] Completed 440000 out of 4000000 steps (11%)

[12:03:18] Timered checkpoint triggered.

[12:18:18] Timered checkpoint triggered.

[12:20:47] Writing local files

[12:20:47] Completed 480000 out of 4000000 steps (12%)

[12:35:47] Timered checkpoint triggered.

[12:50:48] Timered checkpoint triggered.

[12:53:17] Writing local files

[12:53:17] Completed 520000 out of 4000000 steps (13%)

[13:08:18] Timered checkpoint triggered.

[13:23:17] Timered checkpoint triggered.

[13:25:46] Writing local files

[13:25:46] Completed 560000 out of 4000000 steps (14%)

[13:40:46] Timered checkpoint triggered.

[13:55:47] Timered checkpoint triggered.

[13:58:17] Writing local files

[13:58:17] Completed 600000 out of 4000000 steps (15%)

[14:13:17] Timered checkpoint triggered.

[14:28:17] Timered checkpoint triggered.

[14:30:48] Writing local files

[14:30:48] Completed 640000 out of 4000000 steps (16%)

[14:45:47] Timered checkpoint triggered.

[15:00:47] Timered checkpoint triggered.

[15:03:17] Writing local files

[15:03:17] Completed 680000 out of 4000000 steps (17%)

[15:18:17] Timered checkpoint triggered.

[15:33:17] Timered checkpoint triggered.

[15:35:49] Writing local files

[15:35:49] Completed 720000 out of 4000000 steps (18%)

[15:50:49] Timered checkpoint triggered.

[16:05:49] Timered checkpoint triggered.

[16:08:19] Writing local files

[16:08:19] Completed 760000 out of 4000000 steps (19%)

[16:23:19] Timered checkpoint triggered.

[16:38:19] Timered checkpoint triggered.

[16:40:49] Writing local files

[16:40:49] Completed 800000 out of 4000000 steps (20%)

[16:55:48] Timered checkpoint triggered.

[17:10:48] Timered checkpoint triggered.

[17:13:18] Writing local files

[17:13:18] Completed 840000 out of 4000000 steps (21%)

[17:28:18] Timered checkpoint triggered.

[17:37:28] - Autosending finished units...

[17:37:28] Trying to send all finished work units

[17:37:28] + No unsent completed units remaining.

[17:37:28] - Autosend completed

[17:43:19] Timered checkpoint triggered.

[17:45:48] Writing local files

[17:45:48] Completed 880000 out of 4000000 steps (22%)

[18:00:49] Timered checkpoint triggered.

[18:15:49] Timered checkpoint triggered.

[18:18:17] Writing local files

[18:18:17] Completed 920000 out of 4000000 steps (23%)

[18:33:18] Timered checkpoint triggered.

[18:48:17] Timered checkpoint triggered.

[18:50:45] Writing local files

[18:50:45] Completed 960000 out of 4000000 steps (24%)

[19:05:46] Timered checkpoint triggered.

[19:20:46] Timered checkpoint triggered.

[19:23:13] Writing local files

[19:23:13] Completed 1000000 out of 4000000 steps (25%)

[19:38:13] Timered checkpoint triggered.

[19:53:13] Timered checkpoint triggered.

[19:55:43] Writing local files

[19:55:43] Completed 1040000 out of 4000000 steps (26%)

[20:10:42] Timered checkpoint triggered.

[20:25:42] Timered checkpoint triggered.

[20:28:13] Writing local files

[20:28:13] Completed 1080000 out of 4000000 steps (27%)

[20:43:13] Timered checkpoint triggered.

[20:58:13] Timered checkpoint triggered.

[21:00:43] Writing local files

[21:00:43] Completed 1120000 out of 4000000 steps (28%)

[21:15:42] Timered checkpoint triggered.

[21:30:42] Timered checkpoint triggered.

[21:33:13] Writing local files

[21:33:13] Completed 1160000 out of 4000000 steps (29%)

[21:48:13] Timered checkpoint triggered.

[22:03:13] Timered checkpoint triggered.

[22:05:41] Writing local files

[22:05:41] Completed 1200000 out of 4000000 steps (30%)

[22:20:41] Timered checkpoint triggered.

[22:35:41] Timered checkpoint triggered.

[22:38:10] Writing local files

[22:38:10] Completed 1240000 out of 4000000 steps (31%)

[22:53:10] Timered checkpoint triggered.

[23:08:10] Timered checkpoint triggered.

[23:10:40] Writing local files

[23:10:40] Completed 1280000 out of 4000000 steps (32%)

[23:25:40] Timered checkpoint triggered.

[23:37:26] - Autosending finished units...

[23:37:26] Trying to send all finished work units

[23:37:26] + No unsent completed units remaining.

[23:37:26] - Autosend completed

[23:40:40] Timered checkpoint triggered.

[23:43:11] Writing local files

[23:43:11] Completed 1320000 out of 4000000 steps (33%)

[23:58:12] Timered checkpoint triggered.

[00:13:13] Timered checkpoint triggered.

[00:15:40] Writing local files

[00:15:40] Completed 1360000 out of 4000000 steps (34%)

[00:30:41] Timered checkpoint triggered.

[00:45:41] Timered checkpoint triggered.

[00:48:10] Writing local files

[00:48:10] Completed 1400000 out of 4000000 steps (35%)

[01:03:11] Timered checkpoint triggered.

[01:18:10] Timered checkpoint triggered.

[01:20:40] Writing local files

[01:20:40] Completed 1440000 out of 4000000 steps (36%)

[01:35:40] Timered checkpoint triggered.

[01:50:41] Timered checkpoint triggered.

[01:53:10] Writing local files

[01:53:10] Completed 1480000 out of 4000000 steps (37%)

[02:08:11] Timered checkpoint triggered.

[02:18:54] Quit 101 - Fatal error:

[02:18:54] Step 1511662, time 3023.32 (ps) LINCS WARNING

[02:18:54] relative constraint deviation after LINCS:

[02:18:54] max 0.002446 (between atoms 156 and 158) rms 0.000152

[02:18:54]

[02:18:54] Simulation instability has been encountered. The run has entered a

[02:18:54] state from which no further progress can be made.

[02:18:54] This may be the correct result of the simulation, however if you

[02:18:54] often see other project units terminating early like this

[02:18:54] too, you may wish to check the stability of your computer (issues

[02:18:54] such as high temperature, overclocking, etc.).

[02:18:54] Going to send back what have done.

[02:18:54] logfile size: 28565

[02:18:54] - Writing 29248 bytes of core data to disk...

[02:18:54] ... Done.

[02:18:54]

[02:18:54] Folding@home Core Shutdown: EARLY_UNIT_END

[02:18:54] CoreStatus = 72 (114)

[02:18:54] Sending work to server

 

 

[02:18:54] + Attempting to send results

[02:18:54] - Reading file work/wuresults_02.dat from core

[02:18:54] (Read 29248 bytes from disk)

[02:18:54] Connecting to http://171.65.103.160:8080/

[02:18:56] Posted data.

[02:18:56] Initial: 0000; - Uploaded at ~14 kB/s

[02:18:56] - Averaged speed for that direction ~18 kB/s

[02:18:56] + Results successfully sent

[02:18:56] Thank you for your contribution to Folding@Home.

[02:18:56] Trying to send all finished work units

[02:18:56] + No unsent completed units remaining.

[02:18:56] - Preparing to get new work unit...

[02:18:56] + Attempting to get work packet

[02:18:56] - Will indicate memory of 1506 MB.

[02:18:56] - Connecting to assignment server

[02:18:56] Connecting to http://assign.stanford.edu:8080/

[02:18:57] Posted data.

[02:18:57] Initial: 41AB; - Successful: assigned to (171.65.103.160).

[02:18:57] + News From Folding@Home: Welcome to Folding@Home

[02:18:58] Loaded queue successfully.

[02:18:58] Connecting to http://171.65.103.160:8080/

[02:18:59] Posted data.

[02:18:59] Initial: 0000; - Receiving payload (expected size: 398666)

[02:19:05] - Downloaded at ~64 kB/s

[02:19:05] - Averaged speed for that direction ~63 kB/s

[02:19:05] + Received work.

[02:19:05] Trying to send all finished work units

[02:19:05] + No unsent completed units remaining.

[02:19:05] + Closed connections

[02:19:10]

[02:19:10] + Processing work unit

[02:19:10] Core required: FahCore_78.exe

[02:19:10] Core found.

[02:19:10] Working on Unit 03 [september 25 02:19:10]

[02:19:10] + Working ...

[02:19:10] - Calling './FahCore_78.exe -dir work/ -suffix 03 -checkpoint 15 -forceasm -verbose -lifeline 457 -version 502'

 

[02:19:10]

[02:19:10] *------------------------------*

[02:19:10] Folding@Home Gromacs Core

[02:19:10] Version 1.90 (March 8, 2006)

[02:19:10]

[02:19:10] Preparing to commence simulation

[02:19:10] - Assembly optimizations manually forced on.

[02:19:10] - Not checking prior termination.

[02:19:11] - Expanded 398154 -> 2039793 (decompressed 512.3 percent)

[02:19:11] - Starting from initial work packet

[02:19:11]

[02:19:11] Project: 3405 (Run 9, Clone 124, Gen 27)

[02:19:11]

[02:19:11] Assembly optimizations on if available.

[02:19:11] Entering M.D.

[02:19:17] Protein: p3405_dpdp-ii TIP3P ff03

[02:19:17]

[02:19:17] Writing local files

[02:19:17] Extra SSE boost OK.

[02:19:17] Writing local files

[02:19:17] Completed 0 out of 4000000 steps (0%)

 

What's up with that?

 

Står noke om ustabil pc, men det tviler eg på at den er...

[Xell]

Så lenge den sendes inn blir du kreditert for den tiden du har brukt, så du mister ikke noe PPD ved å få slike pakker. Dersom du får I/O error derimot mister du som regel poengene. Siden ting beregnes med noe varierende parametere for å komme frem til forskjellige resultater vil det fra tid til annen skje at ting ikke lar seg beregne ferdig. Dette er også grunnen til at disse resultatene også blir sendt inn. De er også ute etter å se hvilke resultater som ender opp slik. Det med ustabil maskin står vel at hvis dette skjer til stadeighet så kan det skyldes ustabiliteter. Når det ikke skjer til stadighet er det bare en av de tilfeldige ufallene i prosjektet.

[Fruktkake]

Kan noen svare litt kjapt? Har lest FAQ-en om hvilke kort den støtter. Men støtter GPU-klienten dette kortet? Det er timetilbud, så om dere svarer før 18:00 er det fint.