|
Autor
|
Thema: Geschwindigkeitseinbruch durch Cluster? (1324 mal gelesen)
|
RonZ
Mitglied
MB Student
   Beiträge: 15
Registriert: 05.02.2013 VMware Ubuntu 13.04 @
i3-2300 | Win7 und
i7-2670QM | Win7
|
erstellt am: 06. Feb. 2013 02:13  <-- editieren / zitieren -->   Unities abgeben:          
Hallo OF-Gemeinde, ich konnte mir hier im Forum schon einiges an Inspiration anlesen, komme aber mit meinem aktuellen Problem nicht weiter: Nach erfolgreicher Installation und Konfiguration aller notwendigen Werkzeuge für die Cluster-gestützte Berechnung von laminaren, mehrphasigen Strömungen, kam es bei den ersten Tests zu einer eigenartigen Beobachtung: Im Vergleich zur Berechnung des dambreak4phaseFine-Falles jeweils auf nur einem der beiden mir zur Verfügung stehenden Rechnern auf jeweils allen Kernen (siehe System-Info), ist die Rechengeschwindigkeit im Verbund (8 logische Kerne) um Potenzen niedriger. Nachdem auf dem i3 ein timestep mit 2 physischen Kernen im Schnitt 300-500ms dauert, braucht ein Schritt im Duett schonmal gerne eine gute Minute! Was ist denn da los? Hier meine Konfig.:
- 2 Win7 64-bit Maschinen, mittels VMware-Player Ubuntu 12.04 32-bit gebootet, ~4GB zugewiesen
- Verbunden über 1Gbit/s, crossover
- OpenFOAM 2.11 samt abhängigem openMPI 1.4.3, skaliert auf der lokalen Maschine tadellos
- Die Datei 'hosts' wurde entsprechend angepasst, klappt auch. Loginnamen sind identisch und ohne root-Rechte
- ssh mittels rsa-Key authorisiert und passwortloser Login mgl.
- mittels NFS das Arbeitsverzeichnis '~/OpenFOAM' freigegeben und im Slave auf gleichem Pfad gemountet. Habs aber auch schon mit lokalem Root-Verz. auf dem Slave versucht und die processorX-Ordner manuell verteilt. In der decomposeParDict sind 8 Subdomains angegeben, Methodik ist 'simple' und distributed wird nicht. Der decomposedPar-Prozess läuft freilich durch, aber neben
Code:
mpirun -np 8 -machinefile machines multiphaseInterFoam -parallel
bringt auch
Code:
foamJob -s -p multiphaseInterFoam
den besagten Effekt  Laut top hat jeder Rechner auch vier Prozesse, auf Volllast, es funktioniert ja formal auch, nur quälend langsam.
Die Netzwerklast liegt im Bereich von 0,1 und 0,5%. Gut, der case hat nur lächerliche 34200 Zellen, aber das kann doch nicht der Grund sein.Sollte ich eine aktuellere als die mitgelieferte openMPI-Version versuchen? Wenn ja, wie bekomme ich die alte aus ihren ganzen Verknüpfungen raus und implementiere die neue?
Könnte die Virtualisierung das Problem sein?
Gibt es eine Möglichkeit die OF-Config auszuschließen und die Funktion/Skalierung von openMPI mit einer anderen Software zu testen?
Ist es normal, dass beim Übertragen von großen (Test-)Files von VM zu VM die CPU-Last auf beiden Rechnern auf ca. 20% nach oben schnellt (Prozesse: nfsd und cp)? Ich weiß nicht mehr weiter .. Vielen Dank schonmal, dass ihr euch dem bis hierher schonmal angenommen habt!
Freundlichst,
Ronny
Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP |
TTB
Mitglied
CFD Engineer
Beiträge: 353
Registriert: 02.10.2008 BIM HVACTool für Windows OpenFOAM-2.2.x
|
erstellt am: 10. Feb. 2013 17:43 <-- editieren / zitieren --> Unities abgeben: Nur für RonZ 
Hallo, kannst du mal kurz ein paar Schritte vom Log zeigen. Ich würde auf die Kommunikation tippen, obwohl die Netzwerklast im Bereich von 0,1 und 0,5% sei. Viele Grüße
Thomas ------------------
Unsere Software:
HVAC TOOL
Das graphische Interface für OpenFOAM® und TRNSYS Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP |
RonZ
Mitglied
MB Student
Beiträge: 15
Registriert: 05.02.2013 VMware Ubuntu 13.04 @
i3-2300 | Win7 und
i7-2670QM | Win7
|
erstellt am: 10. Feb. 2013 22:52 <-- editieren / zitieren --> Unities abgeben:
Natürlich gern! Erstmal der Lauf auf dem einzelnen DualCore-Rechner mit 4 logischen Kernen bis zum Zeitschritt 0,01s: Code:
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.1.1 |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.1.1-221db2718bbb
Exec : multiphaseInterFoam -parallel
Date : Feb 10 2013
Time : 22:36:01
Host : "silverstone"
PID : 4697
Case : /home/ronz/OpenFOAM/ronz-2.1.1/multiphase/multiphaseInterFoam/laminar/damBreak4phaseFine
nProcs : 4
Slaves :
3
(
"silverstone.4698"
"silverstone.4699"
"silverstone.4700"
)Pstream initialized with:
floatTransfer : 0
nProcsSimpleSum : 0
commsType : nonBlocking
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time Create mesh for time = 0 Reading field p_rgh Reading field U Reading/calculating face flux field phi Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting turbulence model type laminar Reading g
Calculating field g.h
PIMPLE: Operating solver in PISO mode
time step continuity errors : sum local = 0, global = 0, cumulative = 0
GAMGPCG: Solving for pcorr, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 0, global = 0, cumulative = 0
Courant Number mean: 0 max: 0 Starting time loop Courant Number mean: 0 max: 0
Interface Courant Number mean: 0 max: 0
deltaT = 0.00111111
Time = 0.00111111 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
GAMG: Solving for p_rgh, Initial residual = 1, Final residual = 0.045689, No Iterations 1
time step continuity errors : sum local = 0.0441427, global = -1.91015e-06, cumulative = -1.91015e-06
GAMGPCG: Solving for p_rgh, Initial residual = 0.0157164, Final residual = 9.28079e-08, No Iterations 9
time step continuity errors : sum local = 2.62008e-07, global = 2.15607e-08, cumulative = -1.88859e-06
ExecutionTime = 1.05 s ClockTime = 1 s Courant Number mean: 0.00273659 max: 0.182741
Interface Courant Number mean: 0 max: 0
deltaT = 0.00126984
Time = 0.00238095 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1
Phase-sum volume fraction, min, max = 1 0.999995 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00001
Phase-sum volume fraction, min, max = 1 0.999991 1.00001
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00001
Phase-sum volume fraction, min, max = 1 0.999986 1.00001
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00002
Phase-sum volume fraction, min, max = 1 0.999982 1.00002
GAMG: Solving for p_rgh, Initial residual = 0.00135174, Final residual = 3.31298e-05, No Iterations 2
time step continuity errors : sum local = 0.000171422, global = 3.01622e-08, cumulative = -1.85843e-06
GAMGPCG: Solving for p_rgh, Initial residual = 3.49744e-05, Final residual = 7.90431e-08, No Iterations 5
time step continuity errors : sum local = 4.08707e-07, global = 1.98301e-08, cumulative = -1.8386e-06
ExecutionTime = 1.86 s ClockTime = 2 s Courant Number mean: 0.00660949 max: 0.0998965
Interface Courant Number mean: 1.41255e-05 max: 0.0998965
deltaT = 0.00152381
Time = 0.00390476 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00001
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00004
Phase-sum volume fraction, min, max = 1 0.999961 1.00004
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00001
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00006
Phase-sum volume fraction, min, max = 1 0.999922 1.00006
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00008
Phase-sum volume fraction, min, max = 1 0.999884 1.00008
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00011
Phase-sum volume fraction, min, max = 1 0.999846 1.00011
GAMG: Solving for p_rgh, Initial residual = 0.002225, Final residual = 3.37433e-05, No Iterations 2
time step continuity errors : sum local = 0.000241001, global = 3.03352e-07, cumulative = -1.53524e-06
GAMGPCG: Solving for p_rgh, Initial residual = 3.52561e-05, Final residual = 6.80466e-08, No Iterations 5
time step continuity errors : sum local = 4.85085e-07, global = -4.52171e-08, cumulative = -1.58046e-06
ExecutionTime = 2.53 s ClockTime = 2 s Courant Number mean: 0.0129579 max: 0.178755
Interface Courant Number mean: 0.000109037 max: 0.163794
deltaT = 0.00152381
Time = 0.00542857 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 -1.018e-64 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00011
Phase-sum volume fraction, min, max = 1 0.999829 1.00011
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00012
Phase-sum volume fraction, min, max = 1 0.999813 1.00012
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 -1.23797e-40 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00012
Phase-sum volume fraction, min, max = 1 0.999797 1.00012
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 -1.23797e-40 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00012
Phase-sum volume fraction, min, max = 1 0.999781 1.00012
GAMG: Solving for p_rgh, Initial residual = 0.00346581, Final residual = 5.33874e-05, No Iterations 2
time step continuity errors : sum local = 0.000373023, global = 3.28704e-07, cumulative = -1.25176e-06
GAMGPCG: Solving for p_rgh, Initial residual = 5.5663e-05, Final residual = 5.74618e-08, No Iterations 6
time step continuity errors : sum local = 4.0026e-07, global = -3.46379e-08, cumulative = -1.28639e-06
ExecutionTime = 3.16 s ClockTime = 3 s Courant Number mean: 0.0179853 max: 0.229299
Interface Courant Number mean: 0.000187858 max: 0.19852
deltaT = 0.00152381
Time = 0.00695238 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00013
Phase-sum volume fraction, min, max = 1 0.999792 1.00013
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00014
Phase-sum volume fraction, min, max = 1 0.999802 1.00014
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00014
Phase-sum volume fraction, min, max = 1 0.999812 1.00014
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00015
Phase-sum volume fraction, min, max = 1 0.999821 1.00015
GAMG: Solving for p_rgh, Initial residual = 0.004699, Final residual = 7.38092e-05, No Iterations 2
time step continuity errors : sum local = 0.000507482, global = 5.58018e-07, cumulative = -7.28377e-07
GAMGPCG: Solving for p_rgh, Initial residual = 7.7168e-05, Final residual = 6.46996e-08, No Iterations 6
time step continuity errors : sum local = 4.4298e-07, global = -4.1404e-08, cumulative = -7.69781e-07
ExecutionTime = 3.81 s ClockTime = 4 s Courant Number mean: 0.0229823 max: 0.273641
Interface Courant Number mean: 0.000291921 max: 0.235149
deltaT = 0.00152381
Time = 0.00847619 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00015
Phase-sum volume fraction, min, max = 1 0.999834 1.00015
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00016
Phase-sum volume fraction, min, max = 1 0.999846 1.00016
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00016
Phase-sum volume fraction, min, max = 1 0.999858 1.00016
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1.00017
Phase-sum volume fraction, min, max = 1 0.999869 1.00017
GAMG: Solving for p_rgh, Initial residual = 0.00587902, Final residual = 9.39349e-05, No Iterations 2
time step continuity errors : sum local = 0.000636745, global = 6.52213e-07, cumulative = -1.17568e-07
GAMGPCG: Solving for p_rgh, Initial residual = 9.86894e-05, Final residual = 7.19468e-08, No Iterations 6
time step continuity errors : sum local = 4.85126e-07, global = -4.71284e-08, cumulative = -1.64696e-07
ExecutionTime = 4.46 s ClockTime = 4 s Courant Number mean: 0.0279447 max: 0.313225
Interface Courant Number mean: 0.000385222 max: 0.274972
deltaT = 0.00152381
Time = 0.01
Und hier das Bild entstanden mit beiden Maschinen, auch bis T=0,01s:
Code:
/*---------------------------------------------------------------------------*\
| ========= | |
| \\ / F ield | OpenFOAM: The Open Source CFD Toolbox |
| \\ / O peration | Version: 2.1.1 |
| \\ / A nd | Web: www.OpenFOAM.org |
| \\/ M anipulation | |
\*---------------------------------------------------------------------------*/
Build : 2.1.1-221db2718bbb
Exec : multiphaseInterFoam -parallel
Date : Feb 10 2013
Time : 22:27:59
Host : "silverstone"
PID : 3835
Case : /home/ronz/OpenFOAM/ronz-2.1.1/multiphase/multiphaseInterFoam/laminar/damBreak4phaseFine
nProcs : 8
Slaves :
7
(
"silverstone.3836"
"silverstone.3837"
"silverstone.3838"
"vostro.4127"
"vostro.4128"
"vostro.4129"
"vostro.4130"
)Pstream initialized with:
floatTransfer : 0
nProcsSimpleSum : 0
commsType : nonBlocking
sigFpe : Enabling floating point exception trapping (FOAM_SIGFPE).
fileModificationChecking : Monitoring run-time modified files using timeStampMaster
allowSystemOperations : Disallowing user-supplied system call operations // * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //
Create time Create mesh for time = 0 Reading field p_rgh Reading field U Reading/calculating face flux field phi Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting incompressible transport model Newtonian
Selecting turbulence model type laminar Reading g
Calculating field g.h
PIMPLE: Operating solver in PISO mode
time step continuity errors : sum local = 0, global = 0, cumulative = 0
GAMGPCG: Solving for pcorr, Initial residual = 0, Final residual = 0, No Iterations 0
time step continuity errors : sum local = 0, global = 0, cumulative = 0
Courant Number mean: 0 max: 0 Starting time loop Courant Number mean: 0 max: 0
Interface Courant Number mean: 0 max: 0
deltaT = 0.00111111
Time = 0.00111111 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 1 1
GAMG: Solving for p_rgh, Initial residual = 1, Final residual = 0.0457589, No Iterations 1
time step continuity errors : sum local = 0.0442102, global = -2.04911e-06, cumulative = -2.04911e-06
GAMGPCG: Solving for p_rgh, Initial residual = 0.0195481, Final residual = 5.65199e-08, No Iterations 10
time step continuity errors : sum local = 1.28476e-07, global = 4.15075e-09, cumulative = -2.04496e-06
ExecutionTime = 5.79 s ClockTime = 6 s Courant Number mean: 0.00273467 max: 0.182919
Interface Courant Number mean: 0 max: 0
deltaT = 0.00126984
Time = 0.00238095 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749001 0 1
Phase-sum volume fraction, min, max = 1 0.999996 1
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00001
Phase-sum volume fraction, min, max = 1 0.999992 1.00001
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00001
Phase-sum volume fraction, min, max = 1 0.999988 1.00001
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00002
Phase-sum volume fraction, min, max = 1 0.999985 1.00002
GAMG: Solving for p_rgh, Initial residual = 0.00135153, Final residual = 3.23686e-05, No Iterations 2
time step continuity errors : sum local = 0.000167476, global = 7.0186e-08, cumulative = -1.97477e-06
GAMGPCG: Solving for p_rgh, Initial residual = 3.41854e-05, Final residual = 7.25511e-08, No Iterations 5
time step continuity errors : sum local = 3.75202e-07, global = 2.30749e-08, cumulative = -1.9517e-06
ExecutionTime = 8.7 s ClockTime = 9 s Courant Number mean: 0.00661056 max: 0.0998994
Interface Courant Number mean: 1.41308e-05 max: 0.0998994
deltaT = 0.00152381
Time = 0.00390476 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00003
Phase-sum volume fraction, min, max = 1 0.999962 1.00003
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00001
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00005
Phase-sum volume fraction, min, max = 1 0.999923 1.00005
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00001
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00007
Phase-sum volume fraction, min, max = 1 0.999884 1.00008
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.0001
Phase-sum volume fraction, min, max = 1 0.999846 1.0001
GAMG: Solving for p_rgh, Initial residual = 0.002225, Final residual = 3.38309e-05, No Iterations 2
time step continuity errors : sum local = 0.00024162, global = 3.70327e-07, cumulative = -1.58137e-06
GAMGPCG: Solving for p_rgh, Initial residual = 3.5326e-05, Final residual = 5.45711e-08, No Iterations 5
time step continuity errors : sum local = 3.89239e-07, global = 9.43953e-10, cumulative = -1.58043e-06
ExecutionTime = 11.54 s ClockTime = 12 s Courant Number mean: 0.0129571 max: 0.178766
Interface Courant Number mean: 0.000109047 max: 0.163807
deltaT = 0.00152381
Time = 0.00542857 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00011
Phase-sum volume fraction, min, max = 1 0.999837 1.00011
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00011
Phase-sum volume fraction, min, max = 1 0.999827 1.00011
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00012
Phase-sum volume fraction, min, max = 1 0.999818 1.00012
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 -1.43445e-84 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00013
Phase-sum volume fraction, min, max = 1 0.999809 1.00013
GAMG: Solving for p_rgh, Initial residual = 0.00346605, Final residual = 5.34697e-05, No Iterations 2
time step continuity errors : sum local = 0.000373594, global = 4.29735e-07, cumulative = -1.15069e-06
GAMGPCG: Solving for p_rgh, Initial residual = 5.56982e-05, Final residual = 7.79872e-08, No Iterations 5
time step continuity errors : sum local = 5.43781e-07, global = 4.90451e-09, cumulative = -1.14579e-06
ExecutionTime = 14.52 s ClockTime = 15 s Courant Number mean: 0.0179838 max: 0.229313
Interface Courant Number mean: 0.000187866 max: 0.19852
deltaT = 0.00152381
Time = 0.00695238 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00013
Phase-sum volume fraction, min, max = 1 0.999808 1.00013
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00013
Phase-sum volume fraction, min, max = 1 0.999807 1.00013
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00014
Phase-sum volume fraction, min, max = 1 0.999806 1.00014
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 -5.46317e-109 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00014
Phase-sum volume fraction, min, max = 1 0.999805 1.00014
GAMG: Solving for p_rgh, Initial residual = 0.00469895, Final residual = 7.39731e-05, No Iterations 2
time step continuity errors : sum local = 0.000508592, global = 6.15067e-07, cumulative = -5.30719e-07
GAMGPCG: Solving for p_rgh, Initial residual = 7.72245e-05, Final residual = 9.83815e-08, No Iterations 5
time step continuity errors : sum local = 6.7456e-07, global = 1.61615e-08, cumulative = -5.14557e-07
ExecutionTime = 17.26 s ClockTime = 17 s Courant Number mean: 0.0229817 max: 0.273674
Interface Courant Number mean: 0.000291943 max: 0.235199
deltaT = 0.00152381
Time = 0.00847619 MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00016
Phase-sum volume fraction, min, max = 1 0.999807 1.00016
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00018
Phase-sum volume fraction, min, max = 1 0.99981 1.00018
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1.00001
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.0002
Phase-sum volume fraction, min, max = 1 0.999812 1.0002
MULES: Solving for alphawater
water volume fraction, min, max = 0.0831387 0 1
MULES: Solving for alphaoil
oil volume fraction, min, max = 0.1255 0 1.00002
MULES: Solving for alphamercury
mercury volume fraction, min, max = 0.0423611 0 1.00001
MULES: Solving for alphaair
air volume fraction, min, max = 0.749 0 1.00022
Phase-sum volume fraction, min, max = 1 0.999814 1.00022
GAMG: Solving for p_rgh, Initial residual = 0.00587909, Final residual = 9.42588e-05, No Iterations 2
time step continuity errors : sum local = 0.000638915, global = 7.65708e-07, cumulative = 2.51151e-07
GAMGPCG: Solving for p_rgh, Initial residual = 9.88354e-05, Final residual = 2.9003e-08, No Iterations 6
time step continuity errors : sum local = 1.95928e-07, global = 8.56147e-09, cumulative = 2.59712e-07
ExecutionTime = 20.59 s ClockTime = 21 s Courant Number mean: 0.027947 max: 0.313281
Interface Courant Number mean: 0.000385256 max: 0.275048
deltaT = 0.00152381
Time = 0.01
Also gut 5 mal so lange bis zum besagten Zeitschritt. Dabei ist das jetzt schon ertragbar. Ich hab jetzt auf Ubuntu 13.04 gewechselt, weil ich etwas von einem verbuggten NFS-Mechanismus in den letzten Ubuntu-Versionen gelesen habe. Jetzt sind die Übertragungsraten zwar minimalst besser, aber die Latenzen scheinbar noch zu hoch. Die Versuche zu NFS zu optimieren haben nicht wirklich gefruchtet. Die CPU-Last ist immernoch gefühlt zu hoch. Aber was kann man denn da machen? Ich bin doch nicht der erste der NFS dafür nutzen möchte. Gibts da geeignetere Protokolle/Filesystems dafür? Ich bin leider immernoch nicht fündig geworden .. Ich danke dir erstmal für deine Anteilnahme  Grüße
Ronny P.S.: Anbei auch ein Bild vom Systemüberwacher bei einer Dateiübertragung. Die CPU-Last von ~40% bei lächerlichen 10MBit/s kann doch kein normaler Wert sein, oder? [Diese Nachricht wurde von RonZ am 10. Feb. 2013 editiert.] [Diese Nachricht wurde von RonZ am 10. Feb. 2013 editiert.] Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP |
TTB
Mitglied
CFD Engineer
Beiträge: 353
Registriert: 02.10.2008 BIM HVACTool für Windows OpenFOAM-2.2.x
|
erstellt am: 12. Feb. 2013 23:06 <-- editieren / zitieren --> Unities abgeben: Nur für RonZ
Hallo, hmmm, der Unterschied ist ja doch gewaltig. Entweder ist der Fall noch zu klein, damit es sich überhaupt lohnt, parallel zu rechnen oder die Kommunikation werden durch irgendwelche Dienste, Programme? verlangsamt. Evtl. verlangsamt die Sync. von NFS dein System? Umschalten auf Async? Ich werde morgen das Tutorial auf 2 Nodes mal nachrechnen und den Log vergleichen... Viele Grüße
Thomas ------------------
Unsere Software:
HVAC TOOL
Ein graphisches Interface für OpenFOAM® und TRNSYS Eine Antwort auf diesen Beitrag verfassen (mit Zitat/Zitat des Beitrags) IP |
| Anzeige.:
Anzeige: (Infos zum Werbeplatz >>)
 |
Foren auf CAD.de (alle Foren)
| 
