Re: [Yade-users] [Question #692681]: result indeterminism and MPI=OFF
Question #692681 on Yade changed: https://answers.launchpad.net/yade/+question/692681 Status: Answered => Solved Luc OGER confirmed that the question is solved: Dear Jan in my first rapid test, the change to iterPeriod solves my problem thanks Luc -- 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 #692681]: result indeterminism and MPI=OFF
Question #692681 on Yade changed: https://answers.launchpad.net/yade/+question/692681 Status: Open => Answered Jan Stránský proposed the following answer: > PyRunner(command='checkUnbalanced()',realPeriod=2,label='checker'), as already discusses in #1. You have the same code, but because of realPeriod=2, it is very likely that two runs produces different results. realPeriod=2 means running the code every two real seconds, which (depending on random circumstances like running other processes on the computer and slowing Yade) may be at different simulation times. Use iterPeriod or virtPeriod and you should get same results 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 #692681]: result indeterminism and MPI=OFF
Question #692681 on Yade changed: https://answers.launchpad.net/yade/+question/692681 Luc OGER gave more information on the question: with the cmake.. cmake -DENABLE_MPI=OFF -DOpenGL_GL_PREFERENCE=GLVND -DCMAKE_VERBOSE_MAKEFILE=ON -DCMAKE_INSTALL_PREFIX=../install -DPYTHON_VERSION=3 -DDISABLE_SAVE_TEMPS=1 -W ../trunk -- 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 #692681]: result indeterminism and MPI=OFF
Question #692681 on Yade changed:
https://answers.launchpad.net/yade/+question/692681
Status: Needs information => Open
Luc OGER gave more information on the question:
here the cmake command
and the py code
#Result indeterminism
#It is naturally expected that running the same simulation several times will
give exactly the same results: although the computation is done
#with finite precision, round-off errors would be deterministically the same at
every run. While this is true for single-threaded computation
#where exact order of all operations is given by the simulation itself, it is
not true anymore in multi-threaded computation which is described in detail in
later sections.
#The straight-forward manner of parallel processing in explicit DEM is given by
the possibility of treating interactions in arbitrary order.
#Strain and stress is evaluated for each interaction independently, but forces
from interactions have to be summed up. If summation order is also arbitrary
#(in Yade, forces are accumulated for each thread in the order interactions are
processed, then summed together), then the results can be slightly different.
For instance
#(1/10.)+(1/13.)+(1/17.)=0.23574660633484162
#(1/17.)+(1/13.)+(1/10.)=0.23574660633484165
#As forces generated by interactions are assigned to bodies in quasi-random
order, summary force Fi
#on the body can be different between single-threaded and multi-threaded
computations, but also between different runs of multi-threaded computation
with exactly the same parameters.
#Exact thread scheduling by the kernel is not predictable since it depends on
asynchronous events (hardware interrupts) and other unrelated tasks running on
the system;
#and it is thread scheduling that ultimately determines summation order of
force contributions from interactions.
# The components of the batch are:
# 1. table with parameters, one set of parameters per line (ccc.table)
# 2. readParamsFromTable which reads respective line from the parameter file
# 3. the simulation muse be run using yade-batch, not yade
#
# $ yade-batch --job-threads=1 03-oedometric-test.table 03-oedometric-test.py
#
# load parameters from file if run in batch
# default values are used if not run from batch
# gravity deposition in box, showing how to plot and save history of data,
# and how to control the simulation while it is running by calling
# python functions from within the simulation loop
#pas_box theta_max converg_min cover_pack_fraction init_seed friction ratio
#model_type bottom_cover pas_box theta_max converg_min cover_pack_fraction
init_seed friction ratio
#.1 55. .002 0.1 210 0.4 4
readParamsFromTable(model_type=1, packing_fraction=0.5,pas_box=0.5,
theta_max=5.0, nb_cycles=3,converg_min=0.5,nb_layers=10,init_seed=10,
friction=0.15,ratio=1)
# make rMean, rRelFuzz, maxLoad accessible directly as variables later
from yade.params.table import *
# import yade modules that we will use below
from yade import pack, plot, export,math
global ratio,nombre,position_prec,position_init # size ratio between the glued
spheres and the moving ones
global converg_min, init_seed,nombre_moving,z_max # coverage percent for moving
spheres
global i_pas,
box_size,pas_box,rayon,step0,step1,step2,step_precedent,gravity_y,gravity_z,theta_max,theta_max,nom_file,filename_yade,traitement_file,str_Angle,
nb_cycles,numero_cycle,num_Angle
#some parameters passed by batch_table:
batch : friction= 0.5
batch : theta_max = 30.0
batch : pas_box = 0.1
batch : ratio = 3 # size ratio between the glued spheres and the moving
ones
batch : cover_pack_fraction = 0.2 # coverage percent for moving spheres
batch : init_seed=10
batch : packing_fraction = 70./100.
batch : model_type = 1
#some parameters:
shear_modulus = 1e5
poisson_ratio = 0.3
young_modulus = 2*shear_modulus*(1+poisson_ratio)
local_damping = 0.01
viscous_normal = 0.021
viscous_shear = 0.8*viscous_normal
angle = math.atan(friction)
# initialisation coordonnees initiales
#newTable("position_init",600,4) # Create a new table with 5 rows and 3 column
#creating a material (FrictMat):
id_SphereMat=O.materials.append(FrictMat(young=young_modulus,poisson=poisson_ratio,density=2500,frictionAngle=angle,label="glass_beads"))
SphereMat=O.materials[id_SphereMat]
box_size = 1.0
i_pas = 0
step0=0
mask1=0b01
num_Angle= pas_box*i_pas
gravity_y = 9.81*sin( num_Angle*3.14/180.0)
gravity_z = 9.81*cos( num_Angle*3.14/180.0)
O.reset
nom_file=str(model_type)+"_"+str(packing_fraction)+"_"+str(theta_max)+"_"+str(nb_cycles)+"_"+str(nb_layers)+"_"+str(ratio)+'_'+str(friction)+'_'+str(pas_box)+'_'+str(converg_min)+'_'+str(init_seed)
filename_yade=nom_file+'.yade'
rayon = 0.025
height = 5.0*rayon*nb_layers
# create rectangular box from facets
O.bodies.append(geom.facetBox((box_size/2.0,box_size/2.0,height),(box_size/2.0,box_size/2.0,height),wallMask=31))
# create empty sphere packing
#
Re: [Yade-users] [Question #692194]: Body temperature rise higher than thermalBndCondValue
Question #692194 on Yade changed: https://answers.launchpad.net/yade/+question/692194 Jiannan Wang gave more information on the question: Hello Robert, I ran the simulation as it is, the temperature went beyond the boundary value (45) at 155 s and up to 47.8035. The body temperature and the pore pressure curve looks a little bit funny. Body and water temperature in the middle looks like [1] and the pore pressure in the middle looks like [2]. Hope it is OK to attach google link here. Thank you again. Best regards Jiannan [1] https://drive.google.com/file/d/1FiV8S1L-2J4mnV9l9-NWtw6zI2gvZQhM/view?usp=sharing [2 ] https://drive.google.com/file/d/1YEaPS-lSdf5XuWVcfAAacmkF7Xoz6dMp/view?usp=sharing -- 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 #692681]: result indeterminism and MPI=OFF
Question #692681 on Yade changed: https://answers.launchpad.net/yade/+question/692681 Status: Open => Needs information Jan Stránský requested more information: Hello, please provide actutal code, it is very difficult without it. E.g. if you have PyRunner with realPeriod, the results might differ, because the code runs each time in different circumstances. Also, MPI != OpenMP. Do you use -j option for yade? 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
[Yade-users] [Question #692681]: result indeterminism and MPI=OFF
New question #692681 on Yade: https://answers.launchpad.net/yade/+question/692681 Helle, I am trying to 'compare' several runs made with different evolutions of parameters with time, wich implies the same 'initial' behavior , so according to https://yade-dem.org/doc/formulation.html?highlight=indeterminism#result-indeterminism using a single thread version of Yade would give the same results so I have compiled with ENABLE_MPI=OFF and seed defined in the makeCloud identical... BUT I got different results already just after the gravity deposition output after having launched twice the same py code though my batch command. What I have missed to obtain my goal? Luc -- 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 #692194]: Body temperature rise higher than thermalBndCondValue
Question #692194 on Yade changed: https://answers.launchpad.net/yade/+question/692194 Status: Needs information => Open Jiannan Wang gave more information on the question: Hello Robert, Thank you for being so patient. I used the yadedaily (20200829-4169~e627a44~bionic1) in the VritualBox on a Mac. I give 3 cores to the VirtualBox, but I didn't run it parallel. Here is the printAllVersions() output. I will re-run the exact setting again and keep you posted with more detail of how simulation ended. ``` Yade version : 20200829-4169~e627a44~bionic1 Yade features : BoostLog PrecisionDouble Odeint VTK OpenMP GTS GUI-Qt5 CGAL PFVFLOW PFVFLOW LINSOLV MPI TWOPHASEFLOW FEMLIKE GL2PS LBMFLOW THERMAL PotentialParticles PotentialBlocks Yade config dir: ~/.yadedaily Yade precision : 53 bits, 15 decimal places, without mpmath ``` 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| | | 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 18.04.5 LTS Best regards Jiannan Wang -- 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 #692579]: Can't reach isotropic state in 2D biaxial compression
Question #692579 on Yade changed: https://answers.launchpad.net/yade/+question/692579 Jérôme Duriez posted a new comment: As the name says TriaxialStressController is designed for real 3D conditions. Tricking it to 2D configuration is possible but may require a very good knowledge of source code (or hope for other answers) triax.meanStress is always a stress average over three axes in particular: https://gitlab.com/yade- dev/trunk/-/blob/master/pkg/dem/TriaxialStressController.cpp#L314 to 316 -- 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 #692478]: Interrogation about generation of capillary files byMatlab program
Question #692478 on Yade changed: https://answers.launchpad.net/yade/+question/692478 Status: Open => Answered Jérôme Duriez proposed the following answer: Hi Luc, No, there is only one way of calculating that (rather arbitrary) maximum dimensionless suction value [1]. Maximum suction is here the first suction value for which the mean filling angle between two (liquid-bonded and contacting) particles is less than 0.7 degrees. "first" referring to a geometric sequence of suction values, starting from low suction (and filling angle >> 0.7 degrees), and multiplying each time by 1.2 [2] There is a "step" aspect here and I think this explains that non monotonous variation. When one is able to high-jump 0.50 m but fails at 1.5 m, it does not mean that 0.5 m really constitutes his high-jumping maximum level (maybe the guy could jump 0.9 m but he did not get the opportunity to try). In terms of liquid bridges, maybe the (r=1.75;uc* = 26389) reached a mean filling angle = 0.4 degrees, while uc* = 26389/1.2 led to just 0.71 degrees, barely passing the test to multiply uc* by 1.2. Whereas the situation would be different for r=2, with the last filling angle reached being equal to 0.69 => no need to increase uc* further, even though the corresponding liquid bridge is not as tiny as the previous one. As for the consequences on the output files, you may see the M(r=2) file missing more capillary bridge configurations than the M(r=1.75) one, in the high suction regime. (Both files miss capillary configurations anyway since theory says that a non-zero capillary attractive force exists until infinite suction, and that's not the case in both files) Nothing to worry if you do not go in that regime. If you go, YADE should output a message saying that no meniscus is found at a contact, which contradicts the above theory. [1] That "Preliminary C" block at https://gitlab.com/yade-dev/trunk/-/blob/master/examples/capillaryLaplaceYoung/writesCapFile.m#L84 [2] https://gitlab.com/yade-dev/trunk/-/blob/master/examples/capillaryLaplaceYoung/writesCapFile.m#L103 -- 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 #692613]: Order of defining object and engine is causing problems
Question #692613 on Yade changed: https://answers.launchpad.net/yade/+question/692613 Status: Open => Needs information Robert Caulk requested more information: Hello, I am not familiar with cylinderConnections but I can offer some tips to find out the issue. Does the problem occur when you use a single material for everything? Are all 4 of those Laws in your engine list actually necessary? Cheers, Robert -- 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 #692626]: A puzzled problem in HCP packing for JCF model
Question #692626 on Yade changed: https://answers.launchpad.net/yade/+question/692626 Status: Open => Needs information Robert Caulk requested more information: Sorry the question requires a bit too much engagement for me to offer proper help, consider simplifying the question and the script. I think you should start by answering this question: is this unique to JCFpm? i.e. does it also happen with CohFrict or FrictMat? Also, your engine list is bloated, you should not have most of the items there. Try to reduce it to only the pieces that are participating in your simulation. For example, do you have any walls or facets or FrictMats? If no, remove the items associated with them. -- 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 #692315]: Calculation time definition
Question #692315 on Yade changed: https://answers.launchpad.net/yade/+question/692315 Status: Expired => Answered Robert Caulk changed the question status: Changing to answered to avoid launchpad janitor. Maybe someone will answer someday. -- 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 #692194]: Body temperature rise higher than thermalBndCondValue
Question #692194 on Yade changed: https://answers.launchpad.net/yade/+question/692194 Status: Open => Needs information Robert Caulk requested more information: >I ran the code as in [1], but the temperature still went beyond thermalBndCondValue. The code works for me. Which version of yade are you using? Which installation method? At what time/timestep during the simulation does the temperature "go beyond the boundary value"? How far beyond the boundary value does it go? Does it happen in with multi and single core runs? Details :-). -- 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 #692194]: Body temperature rise higher than thermalBndCondValue
Question #692194 on Yade changed: https://answers.launchpad.net/yade/+question/692194 Robert Caulk requested more information: >also seems affect the result Please, add details. -- 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
