Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Jan Stránský posted a new comment: Hello, concerning OP, the problem stated was "it runs very slowly ...I wonder if this is normal?", which was discussed and IMO solved. The "last problem" is not difficult to miss among other information. I suggest to open a new question concerning the meshUpdateInterval value (see also [1], point 5). Cheers Jan [1] https://www.yade-dem.org/wiki/Howtoask -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Answered => Solved Ziyu Wang confirmed that the question is solved: No one seems to have noticed my last problem. I want to mark the problem as solved first. I would appreciate it if someone could see and answer -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Ziyu Wang posted a new comment: Ok,I understand it.Thanks Jan,I'd like to wait for an answer on how to set this value. -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Open => Answered Jan Stránský proposed the following answer: Thanks for information. See Bruno's answer and this line: > "flow" 15000 5159923026.425us 98.68% meaning that the flow engine takes 98.68 % of simulation time. Cheers Jan -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Ziyu Wang gave more information on the question: Hi Bruno, You are right! When I change the value of flow.meshUpdateInterval from 2 to 1000(default value),the running speed from 3/s to about 55/s. So a further question is whether the default value can be used, or how to determine this value (I set it to 2 by referring to the setting in [1]). Thanks for help! [1]https://gitlab.com/yade- dev/trunk/-/blob/master/examples/ThermalEngine/thermoHydroMechanical_coupling.py -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Answered => Open Ziyu Wang is still having a problem: Hello,Jan the following is the information from timing.stats() first stage (quick)-- ForceResetter 4000 75521.451us 0.50% InsertionSortCollider59 196899.836us 1.30% InteractionLoop4000 10281851.285us 67.77% GlobalStiffnessTimeStepper 41 169944.973us 1.12% TriaxialStressController 40003179726.783us 20.96% "flow"00.0us 0.00% "data"4 6129.148us 0.04% PyRunner 4273.363us 0.00% NewtonIntegrator 40001260754.597us 8.31% forces sync4000 8066.228us 0.64% motion integration 40001204713.592us 95.55% sync max vel 4000 8042.751us 0.64% terminate 4000 6417.242us 0.51% TOTAL 160001227239.813us 97.34% TOTAL 15171101.436us 100.00% --second stage(very slow with 15000 steps)--- ForceResetter 15000 330213.906us 0.01% InsertionSortCollider 114 251996.562us 0.00% InteractionLoop 15000 51001932.733us 0.98% GlobalStiffnessTimeStepper 150 741695.019us 0.01% TriaxialStressController 150007243186.493us 0.14% "flow"15000 5159923026.425us 98.68% Position buffer 150014307152.614us 0.08% Triangulating 15001 296565.133us 0.01% Update_Volumes15001 27251839.449us 0.53% Factorize + Solve 15001 2441294624.726us 47.31% compute_Forces15001 221043860.595us 4.28% forces.sync() 15001 35730.031us 0.00% viscous forces15001 31865.789us 0.00% Applying Forces 150012064120.419us 0.04% triangulate + init volumes15001 2464117996.31us 47.75% TOTAL135009 5160443755.066us 100.01% "data" 15 35217.754us 0.00% PyRunner 15902.201us 0.00% NewtonIntegrator 150009604592.545us 0.18% forces sync 150004004912.235us 41.70% motion integration150005391893.156us 56.14% sync max vel 15000 37529.859us 0.39% terminate 15000 27078.136us 0.28% TOTAL 69461413.386us 98.51% TOTAL 5229132763.638us 100.00% I don't understand this very well. Please see if there is any abnormality.. Thanks for kindly help! -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list:
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Needs information => Answered Bruno Chareyre proposed the following answer: > flow.meshUpdateInterval=2 This is an enormous workload, it means to rebuild a mesh, a permeability matrix, and factorize it, every 2 iterations. It can only be very slow. See [1] for a more detailed understanding. Bruno [1] https://www.sciencedirect.com/science/article/abs/pii/S0010465519303340 -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Open => Needs information Jan Stránský requested more information: put the timing code in your script: ### O.timingEnabled=True # before any timing yade.timing.reset() # before 1st stage # the running you want to measure while 1: ... O.run(1000,1) yade.timing.stats() yade.timing.reset() # before 2nd stage flow.dead=0 ... O.run(1,True) # instead of just O.run() yade.timing.stats() ### Cheers Jan -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Needs information => Open Ziyu Wang gave more information on the question: Hello,Jan In fact, I don't know how to deal with my last question, and I don't want to reopen the same question.Anyway, I'm very sorry... The following is the information from printAllVersions() Yade version : 20220208-6309~ebf9df3~bionic1 Yade features : LOGGER USEFUL_ERRORS VTK OPENMP GTS QT5 CGAL PFVFLOW PFVFLOW LINSOLV MPI TWOPHASEFLOW LS_DEM FEMLIKE GL2PS LBMFLOW THERMAL PARTIALSAT POTENTIAL_PARTICLES POTENTIAL_BLOCKS Yade config dir: ~/.yadedaily Yade precision : 53 bits, 15 decimal places, without mpmath, PrecisionDouble ``` Libraries used : | library | cmake| C++ | | - | | --- | | boost | 106501 | 1.65.1 | | cgal | | 4.11| | clp | 1.16.11 | 1.16.11 | | cmake | 3.10.2 | | | coinutils | 2.10.14 | 2.10.14 | | compiler | /usr/bin/c++ 7.5.0 | gcc 7.5.0 | | eigen | 3.3.4| 3.3.4 | | freeglut | 2.8.1| | | gl| | 20190911| | ipython | 5.5.0| | | metis | | 5.1.0 | | mpi | 3.1 | ompi:2.1.1 | | mpi4py| 2.0.0| | | mpmath| 1.0.0| | | openblas | | OpenBLAS 0.2.20| | python| 3.6.9| 3.6.9 | | qglviewer | | 2.6.3 | | qt| | 5.9.5 | | sphinx| 1.6.7-final-0| | | sqlite| | 3.22.0 | | suitesparse | 5.1.2| 5.1.2 | | vtk | 6.3.0| 6.3.0 | ``` Linux version : Ubuntu 18.04.6 LTS Architecture : amd64 Little endian : True - About hardware: Memory 15.6GiB processor AMD® Ryzen 7 4800h with radeon graphics × 16 Graphics card NVIDIA GeForce RTX 2060/PCIe/SSE2 disk 101.7 GB I run the simulation with yadedaily -j 12 script.py(actually no matter how many threads or single thread, the running speed is 3.0/s,So I think it is not normal.) I follow your suggestion and run the timing.stats(),but as follows: In [10]: timing.stats() NameCount TimeRel. time --- ForceResetter 00.0us InsertionSortCollider 00.0us "iloop" 00.0us GlobalStiffnessTimeStepper00.0us TriaxialStressController 00.0us "flow"00.0us "thermal" 00.0us "VTKrec" 00.0us NewtonIntegrator 00.0us "recorder"00.0us PyRunner 00.0us TOTAL 0.0us I do not know what happened..I run another normal script and run timing.stats(), and all the data is still 0.. Thanks for help! -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
Re: [Yade-users] [Question #700587]: My fluid-solid coupling script run very slowly
Question #700587 on Yade changed: https://answers.launchpad.net/yade/+question/700587 Status: Open => Needs information Jan Stránský requested more information: Hello, please do not open the same question twice [1]. Concerning where is "problem", if it is "normal" etc., please first define those terms and provide more information about software (Yade version, OS) and hardware information, how you run the simulation (single/multi thread, ...) etc. With the limited information provided, it may be normal and there may be no problem. You have two stages of the simulation. I suggest investigate each separately. A good start is probably measure times of individual engines using timing module [2]. Please provide the timing.stats() result for both stages. Cheers Jan [1] https://answers.launchpad.net/yade/+question/700285 [2] https://yade-dem.org/doc/yade.timing.html -- You received this question notification because your team yade-users is an answer contact for Yade. ___ Mailing list: https://launchpad.net/~yade-users Post to : yade-users@lists.launchpad.net Unsubscribe : https://launchpad.net/~yade-users More help : https://help.launchpad.net/ListHelp
