[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ESPResSo-users] Increase the lattice boltzmann agrid but got error.

From: Axel Arnold
Subject: Re: [ESPResSo-users] Increase the lattice boltzmann agrid but got error.
Date: Thu, 28 May 2015 23:11:45 +0200


Am 28.05.2015 um 05:44 schrieb Joe Ada <address@hidden>:

I think you are right. After I changed the mesh for p3m from 42 to 8 the error of complex value is gone. I guess, I have to change the p3m parameters every time I change the agrid.

In fact, it is probably more safe to let the automatic tuning do its work, including the mesh. The rule of thumb is that there should be around 1 particle per mesh point, so 42 is way too large. This indeed can lead to strange behavior like sharp contrasts in the charge distribution, which leads to numerical errors. So, better rely on "p3m tune accuracy 0.01" or so.

Also, could any one give me some clue about setting the skin? 

The skin determines when the Verlet lists are rebuild and particles are exchanged between processors. A large skin means less communication and calculation of Verlet lists, but more comparisons in the inner particle loop. The optimal value there is a trade off that depends on your hardware (network latency vs. computing power). Fortunately, the computation time as function of the skin typically has a broad minimum, so that you don’t need to excessively tune. As a rule of thumb, the skin should be about 40% of the primary particle diameter. As I said, there are some side constraints: skin must be smaller than agrid/2, and processor domain size minus skin limits the range of short ranged interactions. Because P3M with very small grids tends to use rather large cutoffs, you can hit that limit. However, that was definitely not your case with 42...


On Wed, May 27, 2015 at 1:23 PM, Axel Arnold <address@hidden> wrote:

Just to add some more information:

The electrostatics algorithm used internally a complex FFT on real data, so that the result should be real as well. The code complains that it isn't, though the deviation is small. That typically means that there are strange charge distributions or parameters involved. If the simulation doesn't crash, then it is more likely a parameter problem (a finer agrid requires a smaller skin which might confuse the FFT).


Am 27.05.2015 um 10:05 schrieb Florian Weik <address@hidden>:

this is an error message from the P3M (Electrostatics). This is an indication that something else is wrong.
It would be easier to help you if you would post a minimal example that reproduces you problem. In principle
you should be able to change agrid.


On Wed, May 27, 2015 at 3:30 AM, Joe Ada <address@hidden> wrote:
Hi Guys,

I found that when I increase the agrid of LB fluidics, without changing anything else, I got the following error:
Complex value is not zero (i=16774,data="" class="">
I wanted to increase the size of the simulation cube. Because of the limit of memory, I can only increase agrid. Is there a upper limit of it?


Florian Weik, Dipl.-Phys.,
Institut für Computerphysik, Allmandring 3, D-70569 Stuttgart
Phone: +49-711-685-67703
Public Key 0x0562F11D Fingerprint 3294 6360 EC93 37A3 BD40  F597 0BAD 3AF8 0562 F11D


Axel Arnold
Martha-Schmidtmann-Str. 7
70374 Stuttgart, Germany
Email: address@hidden
Phone: +49 173 870 6659

reply via email to

[Prev in Thread] Current Thread [Next in Thread]