Hello,
I'm running into a problem where my code exits early without giving me any obvious error messages. I'm simply trying to pour particles into my hopper. I'm submitting jobs batch style to a shared computer network.
I have the fix/insert set to use massrate. If I set the mass rate too low, obviously all the particles won't be generated before I run out of timesteps. However, if I set the massrate too high, for some reason the job will stop in the middle of execution. In this case, it stops after the second attempt to insert particles - step 1001.
Is there some sort of magic number for particle insertion rate? Some ratio between insertion rate and step size?
See below for my code and an output file. I know the output file is kind of long and busy, but it makes it easy to tell exactly when the code runs into problems. At the very bottom of the output file, you can see the resources I requested and the resources actually used. "Walltime" is literally the time used according to the clock on the wall, as opposed to CPU time. I've used different parameters to get this job to run successfully, and it can easily complete it in the 10 minute window I've asked for.
Any ideas?
| Attachment | Size |
|---|---|
 nano.save_.2.txt | 2.06 KB |
| standard_out_file.1640774.repace.pace_.gatech.edu_.txt | 9.88 KB |
msandli | Wed, 08/29/2012 - 04:55
I just remembered
that the massrate option is a mass rate per timestep. is there some sort of maximum # of particles per time step the code can insert? I know the number itself would probably depend on the geometry, but is there some sort of upper limit?
ckloss | Wed, 08/29/2012 - 08:33
Hi msandli, there was a bug
Hi msandli,
there was a bug related to that - please update to the current version
Cheers, Christoph
msandli | Fri, 08/31/2012 - 01:55
for now
I'm hoping I can get the updated version, but as I don't handle the software upgrades, I have no timeline. In the mean time, can you describe the bug to me? Or at least ways to avoid it? My IT support is telling me something about memory referencing, but it's all over my head at this point.
also, does this have anything to do with another problem I'm having where I dump some particles in a hopper, let them settle, and then unfix a part at the bottom of the hopper to simulate opening a valve and letting the particles drain out, but the very timestep where I unfix the valve, the program does the same thing and exits without any warning?
Sorry for the constant questions, but I don't know anyone else that uses LIGGGHTS or even LAMMPS on a regular basis where I'm at, so you guys are my only resource.
rahulsoni | Thu, 02/27/2014 - 15:12
units
I am running a simulation with fix insert/stream command. Following is the command in script.
fix insertion particles insert/stream seed 5330 distributiontemplate pdd1
& verbose yes maxattempt 100 mass INF massrate 1.0 start 1000 overlapcheck yes vel constant 0. 0. -1.0 & insertion_face inface extrude_length 0.4
I have assigned massrate = 1.0 but I really don't know how much actually it is. Simulation script is in SI units but from results I do not get mass insetion any way near to 1.0 kg/second or something like that. Please tell me how much actually it is.....