Linux SMP Problems

[mpapad]

Doc, βάλε gpu client... απλά σπέρνει...

(Μου έκαψε ήδη ένα τροφοδοτικό η 9800 GTX αλλά πραγματικά πετάει).

Off Topic

Πετάει τόσο που λέω στο μηχανάκι με την 8800gt να πάρω και άλλες 2 να φολντάρουν και να βγάλω την SMP.

[Iannis]

Off Topic

Πετάει τόσο που λέω στο μηχανάκι με την 8800gt να πάρω και άλλες 2 να φολντάρουν και να βγάλω την SMP.

Να και η επερχομενη nVidia Killer αν και ηδη το cost/benefit με τις νεες 4850-70 ειναι σαφώ υπερ της ΑΤΙ

http://www.xtremesystems.org/forums/...d.php?t=192829

........Auto merged post: Iannis πρόσθεσε 33 λεπτά και 32 δευτερόλεπτα αργότερα ........

απο φασαρια τι λενε?

[mpapad]

Off Topic

Να και η επερχομενη nVidia Killer αν και ηδη το cost/benefit με τις νεες 4850-70 ειναι σαφώ υπερ της ΑΤΙ

http://www.xtremesystems.org/forums/...d.php?t=192829

Δε τσιμπάω. με 130 ευρώπουλα παίρνω 8800gt και μου δίνει 5000 ppd.
H 4850 πόσα ppd δίνει? 2000 απο ότι διαβάζω γιατί λεει θέλει ειδική πρωτείνη για πάρτη της.

απο φασαρια τι λενε?

Ποιές? οι 4850 ή οι 8800?
Για 8800 ούτε που τις ακούω.

Καλύτερα να συνεχίσουμε σε άλλο thread αυτή τη συζήτηση.

[alexopth69]

εβαλα sabayon 3.5 από live και κάνει τα ίδια, 70% καθυστέρηση ο 2.6.25 σε σχέση με το 2.6.24. Γύρισα πίσω σε 2.6.24

[tedd]

Μετά απο πειράματα σε sabyon 3.5 (core 2.6.25) και ubuntu 8.04.1 (core 2.6.24) βρήκα οτι το ubuntu ειναι πιο γρήγορο κατα ~2% στις ίδιες πρωτεινες το τέστ έγιne στο ίδιο μηχανημα με τα ιδια settings και μέσω vmware.
Tελικά η φήμη οτι το ubuntu ειναι πιο αργο στο folding καταρίπτεται. MythBusted
Πάντως το sabyon 3.5 final (core 2.6.25) ειναι ελαφρως πιο γρήγορο κατα ~1% σε σχέση με την προηγούμενη βετα εκδοση (core 2.6.23, αν θυμάμαι καλά), τα πειράματα συνεχίζονται.

[Hengeo]

Εγώ (Ubuntu 8.04, Kernel 2.6.24-19) δεν έχω παράπονο με την ταχύτητα, αλλά μου την έχει σπάσει λίγο το συνεχές 'ντάντεμα' που θέλει. Την μία κολλάει ο server από τον οποίο παίρνει νέες πρωτείνες, την άλλη κολλάει το queue, και πρέπει να κάνεις qfix για να στείλει. Και εντάξει, εγώ έχω ένα pc, και παλεύεται, στις περισσότερες περιπτώσεις το ξεκολλάω σχετικά γρήγορα, αλλά όσοι έχετε πολλά pc, πως την παλεύετε, και δεν σας έχουν σπάσει εντελώς τα νεύρα;

[apollokk]

προβλημα. εβαλα το linux smp client στο arch. και παιρνω αυτο

[kostis@apollo folding]$  ./fah
[kostis@apollo folding]$ 
Note: Please read the license agreement (fah6 -license). Further 
use of this software requires that you have read and accepted this agreement.

2 cores detected


--- Opening Log file [September 4 09:54:16] 


# SMP Client ##################################################################
###############################################################################

                       Folding@Home Client Version 6.02

                          http://folding.stanford.edu

###############################################################################
###############################################################################

Launch directory: /home/kostis/folding
Executable: ./fah6
Arguments: -smp -verbosity 9 

[09:54:16] - Ask before connecting: No
[09:54:16] - User name: apollokk (Team 36673)
[09:54:16] - User ID: 7D83A717F57DB12
[09:54:16] - Machine ID: 1
[09:54:16] 
[09:54:16] Work directory not found. Creating...
[09:54:16] Could not open work queue, generating new queue...
[09:54:16] - Preparing to get new work unit...
[09:54:16] + Attempting to get work packet
[09:54:16] - Will indicate memory of 2011 MB
[09:54:16] - Detect CPU. Vendor: GenuineIntel, Family: 6, Model: 15, Stepping: 13
[09:54:16] - Connecting to assignment server
[09:54:16] Connecting to http://assign.stanford.edu:8080/
[09:54:16] - Autosending finished units...
[09:54:16] Trying to send all finished work units
[09:54:16] + No unsent completed units remaining.
[09:54:16] - Autosend completed
[09:54:16] Posted data.
[09:54:16] Initial: 40AB; - Successful: assigned to (171.64.65.64).
[09:54:16] + News From Folding@Home: Welcome to Folding@Home
[09:54:17] Loaded queue successfully.
[09:54:17] Connecting to http://171.64.65.64:8080/
[09:54:23] Posted data.
[09:54:23] Initial: 0000; - Receiving payload (expected size: 4730926)
[09:54:50] - Downloaded at ~171 kB/s
[09:54:50] - Averaged speed for that direction ~171 kB/s
[09:54:50] + Received work.
[09:54:50] + Closed connections
[09:54:50] 
[09:54:50] + Processing work unit
[09:54:50] Core required: FahCore_a1.exe
[09:54:50] Core not found.
[09:54:50] - Core is not present or corrupted.
[09:54:50] - Attempting to download new core...
[09:54:50] + Downloading new core: FahCore_a1.exe
[09:54:50] Downloading core (/~pande/Linux/x86/Core_a1.fah from www.stanford.edu)
[09:54:50] Initial: AFDE; + 10240 bytes downloaded
[09:54:51] Initial: B54E; + 20480 bytes downloaded
[09:54:51] Initial: D6C2; + 30720 bytes downloaded
[09:54:51] Initial: 9F08; + 40960 bytes downloaded
[09:54:51] Initial: C6C3; + 51200 bytes downloaded
[09:54:51] Initial: EBA8; + 61440 bytes downloaded
[09:54:51] Initial: 3141; + 71680 bytes downloaded
[09:54:51] Initial: D218; + 81920 bytes downloaded
[09:54:51] Initial: F7AC; + 92160 bytes downloaded
[09:54:51] Initial: 820B; + 102400 bytes downloaded
[09:54:51] Initial: 1B1E; + 112640 bytes downloaded
[09:54:51] Initial: C249; + 122880 bytes downloaded
[09:54:52] Initial: 5EBD; + 133120 bytes downloaded
[09:54:52] Initial: CD6C; + 143360 bytes downloaded
[09:54:52] Initial: 221C; + 153600 bytes downloaded
[09:54:52] Initial: DB18; + 163840 bytes downloaded
[09:54:52] Initial: 237E; + 174080 bytes downloaded
[09:54:52] Initial: AEEC; + 184320 bytes downloaded
[09:54:52] Initial: 4C66; + 194560 bytes downloaded
[09:54:52] Initial: AE1E; + 204800 bytes downloaded
[09:54:52] Initial: A37E; + 215040 bytes downloaded
[09:54:52] Initial: 8193; + 225280 bytes downloaded
[09:54:52] Initial: 9F05; + 235520 bytes downloaded
[09:54:52] Initial: AAA5; + 245760 bytes downloaded
[09:54:52] Initial: 6400; + 256000 bytes downloaded
[09:54:52] Initial: 6E3D; + 266240 bytes downloaded
[09:54:52] Initial: EA6B; + 276480 bytes downloaded
[09:54:52] Initial: 820A; + 286720 bytes downloaded
[09:54:52] Initial: DE6D; + 296960 bytes downloaded
[09:54:52] Initial: B97B; + 307200 bytes downloaded
[09:54:52] Initial: 9D5D; + 317440 bytes downloaded
[09:54:52] Initial: 91D7; + 327680 bytes downloaded
[09:54:52] Initial: BB3B; + 337920 bytes downloaded
[09:54:52] Initial: 611B; + 348160 bytes downloaded
[09:54:52] Initial: B290; + 358400 bytes downloaded
[09:54:52] Initial: B0AA; + 368640 bytes downloaded
[09:54:52] Initial: 6A85; + 378880 bytes downloaded
[09:54:52] Initial: BF10; + 389120 bytes downloaded
[09:54:52] Initial: A818; + 399360 bytes downloaded
[09:54:52] Initial: 90E1; + 409600 bytes downloaded
[09:54:52] Initial: 2869; + 419840 bytes downloaded
[09:54:52] Initial: CAFE; + 430080 bytes downloaded
[09:54:52] Initial: 414B; + 440320 bytes downloaded
[09:54:52] Initial: 9B7A; + 450560 bytes downloaded
[09:54:52] Initial: 33AA; + 460800 bytes downloaded
[09:54:52] Initial: B1D5; + 471040 bytes downloaded
[09:54:52] Initial: 0206; + 481280 bytes downloaded
[09:54:52] Initial: 11F4; + 491520 bytes downloaded
[09:54:52] Initial: 31B5; + 501760 bytes downloaded
[09:54:52] Initial: 46B2; + 512000 bytes downloaded
[09:54:52] Initial: 3113; + 522240 bytes downloaded
[09:54:52] Initial: 525A; + 532480 bytes downloaded
[09:54:52] Initial: 66F9; + 542720 bytes downloaded
[09:54:52] Initial: 9672; + 552960 bytes downloaded
[09:54:52] Initial: 9058; + 563200 bytes downloaded
[09:54:52] Initial: 49ED; + 573440 bytes downloaded
[09:54:52] Initial: 515D; + 583680 bytes downloaded
[09:54:53] Initial: CAC0; + 593920 bytes downloaded
[09:54:53] Initial: 0B15; + 604160 bytes downloaded
[09:54:53] Initial: 5A89; + 614400 bytes downloaded
[09:54:53] Initial: 0F31; + 624640 bytes downloaded
[09:54:53] Initial: 2BC3; + 634880 bytes downloaded
[09:54:53] Initial: 3C06; + 645120 bytes downloaded
[09:54:53] Initial: 89C7; + 655360 bytes downloaded
[09:54:53] Initial: 6C54; + 665600 bytes downloaded
[09:54:53] Initial: 8D4D; + 675840 bytes downloaded
[09:54:53] Initial: EA59; + 686080 bytes downloaded
[09:54:53] Initial: C563; + 696320 bytes downloaded
[09:54:53] Initial: 8D45; + 706560 bytes downloaded
[09:54:53] Initial: 9BD0; + 716800 bytes downloaded
[09:54:53] Initial: 130C; + 727040 bytes downloaded
[09:54:53] Initial: CDA1; + 737280 bytes downloaded
[09:54:53] Initial: 7681; + 747520 bytes downloaded
[09:54:53] Initial: 1110; + 757760 bytes downloaded
[09:54:53] Initial: EE35; + 768000 bytes downloaded
[09:54:53] Initial: E5E1; + 778240 bytes downloaded
[09:54:53] Initial: 4B97; + 788480 bytes downloaded
[09:54:53] Initial: 4D75; + 798720 bytes downloaded
[09:54:53] Initial: E268; + 808960 bytes downloaded
[09:54:53] Initial: FAC6; + 819200 bytes downloaded
[09:54:53] Initial: A625; + 829440 bytes downloaded
[09:54:53] Initial: A12A; + 839680 bytes downloaded
[09:54:53] Initial: 83A3; + 849920 bytes downloaded
[09:54:53] Initial: 3BEA; + 860160 bytes downloaded
[09:54:53] Initial: 5298; + 870400 bytes downloaded
[09:54:53] Initial: 4811; + 880640 bytes downloaded
[09:54:53] Initial: EB07; + 890880 bytes downloaded
[09:54:53] Initial: 83FC; + 901120 bytes downloaded
[09:54:53] Initial: FA4E; + 911360 bytes downloaded
[09:54:53] Initial: 2945; + 921600 bytes downloaded
[09:54:53] Initial: 6BC9; + 931840 bytes downloaded
[09:54:53] Initial: E495; + 942080 bytes downloaded
[09:54:53] Initial: 1050; + 952320 bytes downloaded
[09:54:53] Initial: 2070; + 962560 bytes downloaded
[09:54:53] Initial: 1083; + 972800 bytes downloaded
[09:54:53] Initial: 96E5; + 983040 bytes downloaded
[09:54:53] Initial: 3EEE; + 993280 bytes downloaded
[09:54:53] Initial: 84AC; + 1003520 bytes downloaded
[09:54:53] Initial: 3B6B; + 1013760 bytes downloaded
[09:54:53] Initial: 3030; + 1024000 bytes downloaded
[09:54:53] Initial: 4B95; + 1034240 bytes downloaded
[09:54:53] Initial: D9BC; + 1044480 bytes downloaded
[09:54:53] Initial: C5B8; + 1054720 bytes downloaded
[09:54:53] Initial: A5EF; + 1064960 bytes downloaded
[09:54:53] Initial: 28DC; + 1075200 bytes downloaded
[09:54:53] Initial: 0943; + 1085440 bytes downloaded
[09:54:53] Initial: 338A; + 1095680 bytes downloaded
[09:54:53] Initial: ADFC; + 1105920 bytes downloaded
[09:54:54] Initial: ED39; + 1116160 bytes downloaded
[09:54:54] Initial: D284; + 1126400 bytes downloaded
[09:54:54] Initial: 0057; + 1136640 bytes downloaded
[09:54:54] Initial: 3E65; + 1146880 bytes downloaded
[09:54:54] Initial: FCB5; + 1157120 bytes downloaded
[09:54:54] Initial: A7D8; + 1167360 bytes downloaded
[09:54:54] Initial: A564; + 1177600 bytes downloaded
[09:54:54] Initial: 7654; + 1187840 bytes downloaded
[09:54:54] Initial: 0848; + 1198080 bytes downloaded
[09:54:54] Initial: 471E; + 1208320 bytes downloaded
[09:54:54] Initial: A7F3; + 1218560 bytes downloaded
[09:54:54] Initial: FA59; + 1228800 bytes downloaded
[09:54:54] Initial: FBF2; + 1239040 bytes downloaded
[09:54:54] Initial: F54E; + 1249280 bytes downloaded
[09:54:54] Initial: 3023; + 1259520 bytes downloaded
[09:54:54] Initial: AB37; + 1269760 bytes downloaded
[09:54:54] Initial: 0896; + 1280000 bytes downloaded
[09:54:54] Initial: 756D; + 1290240 bytes downloaded
[09:54:54] Initial: C1E7; + 1300480 bytes downloaded
[09:54:54] Initial: 9AAC; + 1310720 bytes downloaded
[09:54:54] Initial: E5AF; + 1320960 bytes downloaded
[09:54:54] Initial: BBE3; + 1331200 bytes downloaded
[09:54:54] Initial: 3596; + 1341440 bytes downloaded
[09:54:54] Initial: 924C; + 1351680 bytes downloaded
[09:54:54] Initial: 30B7; + 1361920 bytes downloaded
[09:54:54] Initial: AEB7; + 1372160 bytes downloaded
[09:54:54] Initial: 7D25; + 1382400 bytes downloaded
[09:54:54] Initial: 0FEB; + 1392640 bytes downloaded
[09:54:54] Initial: 3131; + 1402880 bytes downloaded
[09:54:54] Initial: 755F; + 1413120 bytes downloaded
[09:54:54] Initial: 4800; + 1423360 bytes downloaded
[09:54:54] Initial: 1282; + 1433600 bytes downloaded
[09:54:54] Initial: B2A3; + 1443840 bytes downloaded
[09:54:54] Initial: 21E9; + 1454080 bytes downloaded
[09:54:54] Initial: 789E; + 1464320 bytes downloaded
[09:54:54] Initial: 8542; + 1474560 bytes downloaded
[09:54:54] Initial: 3A56; + 1484800 bytes downloaded
[09:54:54] Initial: D4FE; + 1490945 bytes downloaded
[09:54:54] Verifying core Core_a1.fah...
[09:54:54] Signature is VALID
[09:54:54] 
[09:54:54] Trying to unzip core FahCore_a1.exe
[09:54:55] Decompressed FahCore_a1.exe (3625104 bytes) successfully
[09:54:55] + Core successfully engaged
[09:55:00] 
[09:55:00] + Processing work unit
[09:55:00] Core required: FahCore_a1.exe
[09:55:00] Core found.
[09:55:00] Working on Unit 01 [September 4 09:55:00]
[09:55:00] + Working ...
[09:55:00] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a1.exe -dir work/ -suffix 01 -checkpoint 15 -verbose -lifeline 4450 -version 602'

[09:55:00] 
[09:55:00] *------------------------------*
[09:55:00] Folding@Home Gromacs SMP Core
[09:55:00] Version 1.74 (November 27, 2006)
[09:55:00] 
[09:55:00] Preparing to commence simulation
[09:55:00] - Ensuring status. Please wait.
[09:55:01] - Starting from initial work packet
[09:55:01] 
[09:55:01] Project: 2665 (Run 0, Clone 911, Gen 46)
[09:55:01] 
[09:55:01] Assembly optimizations on if available.
[09:55:01] Entering M.D.
[09:55:18] al work packet
[09:55:18] 
[09:55:18] Project: 2665 (Run 0, Clone 911, Gen 46)
[09:55:18] 
[09:55:19] 65 (Run 0, Clone 911, Gen 46)
[09:55:19] 
[09:55:19] Entering M.D.
NNODES=4, MYRANK=1, HOSTNAME=apollo
NNODES=4, MYRANK=2, HOSTNAME=apollo
NNODES=4, MYRANK=3, HOSTNAME=apollo
NNODES=4, MYRANK=0, HOSTNAME=apollo

τι να κανω;

[vagskarm]

Ακόμη έτσι είναι ? Τσέκαρε αν το hostname που βλεπεις στο log κάτω μέρος είναι δηλωμένο και στο /etc/hosts

[apollokk]

Ακόμη έτσι είναι ? Τσέκαρε αν το hostname που βλεπεις στο log κάτω μέρος είναι δηλωμένο και στο /etc/hosts

ειναι....

[tedd]

Δεν καταλαβαίνω, πού είναι το προβλημα που φαίνεται φυσιολογικό και σε εμένα έτσι είναι ή ειναι κάτι που δεν το προσέχω;

[alexopth69]

αν κολλαει στο σημείο που μας δίνεις το log κοίτα μήπως τρέχει κι άλλο fah process

[Hengeo]

Έχετε και άλλοι πρόβλημα με τις 266χ πρωτεΐνες και τα chekpoint; Ακόμα και αν κλείσω ομαλά την κονσόλα με ctrl-c, όταν την ξαναανοίξω, γίνεται το παρακάτω:

Spoiler:

[08:37:27] - Ask before connecting: No
[08:37:27] - User name: Hengeo (Team 36673)
[08:37:27] - User ID: A5F9CB46F929D47
[08:37:27] - Machine ID: 1
[08:37:27]
[08:37:27] Loaded queue successfully.
[08:37:27] - Autosending finished units...
[08:37:27] Trying to send all finished work units
[08:37:27] + No unsent completed units remaining.
[08:37:27] - Autosend completed
[08:37:27]
[08:37:27] + Processing work unit
[08:37:27] Core required: FahCore_a2.exe
[08:37:27] Core found.
[08:37:27] Working on Unit 01 [September 12 08:37:27]
[08:37:27] + Working ...
[08:37:27] - Calling './mpiexec -np 4 -host 127.0.0.1 ./FahCore_a2.exe -dir work/ -suffix 01 -checkpoint 15 -forceasm -verbose -lifeline 7814 -version 602'

[08:37:27]
[08:37:27] *------------------------------*
[08:37:27] Folding@Home Gromacs SMP Core
[08:37:27] Version 1.91 (2007)
[08:37:27]
[08:37:27] Preparing to commence simulation
[08:37:27] - Ensuring status. Please wait.
[08:37:44] - Assembly optimizations manually forced on.
[08:37:44] - Not checking prior termination.
[08:37:44] Error: Work unit read from disk is invalid
[08:37:44] Finalizing output
[08:37:47] - Expanded 4828034 -> 23976217 (decompressed 496.6 percent)
[08:37:47]
[08:37:47] Project: 2669 (Run 8, Clone 116, Gen 6)
[08:37:47]
[08:37:48] Assembly optimizations on if available.
[08:37:48] Entering M.D.
NNODES=4, MYRANK=1, HOSTNAME=George
NNODES=4, MYRANK=3, HOSTNAME=George
NNODES=4, MYRANK=2, HOSTNAME=George
NNODES=4, MYRANK=0, HOSTNAME=George
NODEID=2 argc=14
NODEID=0 argc=14
NODEID=3 argc=14
NODEID=1 argc=14
:-) G R O M A C S (-:

Groningen Machine for Chemical Simulation

:-) VERSION 3.3.99_development_20070720 (-:


Written by David van der Spoel, Erik Lindahl, Berk Hess, and others.
Copyright (c) 1991-2000, University of Groningen, The Netherlands.
Copyright (c) 2001-2006, The GROMACS development team,
check out http://www.gromacs.org for more information.


:-) mdrun (-:

Reading file work/wudata_01.tpr, VERSION 3.3.99_development_20070618 (single precision)
[08:37:54] Will resume from checkpoint file
starting mdrun '22860 system'
250000 steps, 500.0 ps.
[08:37:56] Resuming from checkpoint
old size= 43437 old_crc=2553054
-------------------------------------------------------
Program mdrun, VERSION 3.3.99_development_20070720
Source code file: md.c, line: 831

Fatal error:
Checkpoint error on step 1558760

-------------------------------------------------------

Thanx for Using GROMACS - Have a Nice Day

Error on node 0, will try to stop all the nodes
Halting parallel program mdrun on CPU 0 out of 4

gcq#0: Thanx for Using GROMACS - Have a Nice Day

[cli_0]: aborting job:
application called MPI_Abort(MPI_COMM_WORLD, -1) - process 0
[cli_1]: aborting job:
Fatal error in MPI_Wait: Error message texts are not available
[08:37:56] File work/wudata_01.log has changed since last checkpoint
[0]0:Return code = 255
[0]1:Return code = 1
[0]2:Return code = 0, signaled with Quit
[0]3:Return code = 0, signaled with Quit
[08:38:01] CoreStatus = FF (255)
[08:38:01] Client-core communications error: ERROR 0xff
[08:38:01] Deleting current work unit & continuing...
After shutdown
After shutdown
After shutdown
[0]0:Return code = 0, signaled with Quit
[0]2:Return code = 0, signaled with Quit
[0]3:Return code = 0, signaled with Quit
[08:42:23] - Warning: Could not delete all work unit files (1): Core file absent
[08:42:23] Trying to send all finished work units
[08:42:23] + No unsent completed units remaining.
[08:42:23] - Preparing to get new work unit...

Με τις 266χ και μόνο γίνεται αυτό, με 2605 δεν έχει πρόβλημα. Δυστυχώς 2-3 φορές που έπρεπε οπωσδήποτε να μπω windows, έχω χάσει αρκετή δουλειά με αυτό το πρόβλημα, αλλά δεν γίνεται αλλιώς..

[mpapad]

Δεν έχω ιδέα από το θέμα στο linux αλλά μήπως δεν κλείνουν τα FahCore όταν πατάς Ctrl-C?

Στα windows το έχω αυτό το θέμα και αν δεν κλείσω εγώ τα FahCore κάνει corrupt το αρχείο.

[vagskarm]

Οταν κόβεις το parent process τότε όλα τα processes που έχουν ξεκινήσει αυτό, σταματάνε κι αυτά.