QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Zexter on November 30, 2010, 18:58

Title: electrostatic difference potential
Post by: Zexter on November 30, 2010, 18:58
Hi
 

when I try to compute the electrostatic difference potential of a Dwnt the vnl terminates automatically after running for an hour.what could be the possible reason
Title: Re: electrostatic difference potential
Post by: ugglebot on November 30, 2010, 19:31
Maybe a virus attack ??? I'm joking, my friend, I'd love to help you figure it out, but it's pretty impossible without at least an error message, the log file, and ideally the input file too :)
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 06:39
Sorry for it


The error displayed is "This application has requested the runtime to terminate it in an unusual way.Please contactr the application's support teanm for more information". After running for an hour.

I was actually trying to calculate the Electrostatic Diffrerence potential and the I stumble upon this error in the log script.My calculator parameters are default values except for the K point sampling 1,1,5

is there any possiblity for the error to lie in the k pooint sampling. I was actually intended to find the inter wall interaction energy,potential relief function of a Double walled nanotube(5,5)/(10,10) case with repetition of 10 in Z axis and 600 atoms count.
Title: Re: electrostatic difference potential
Post by: zh on December 1, 2010, 09:27
It seems that your problem isn't related to the setup of k-point sampling. Just from the brief description of the setup in your calculation, I think that the possible reasons for your problem may come from the followings:
i) in a certain region the interior tube may be too close to the outer one;
ii) the setup parameters specified in your script file may result in the requirment of huge memory, and the job runned out of the memory of your computer.

To figure out the exact reasons, we may need your script file and the corresponding log file.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 12:19
Hi

I am attaching the script file here
I checked the tube configurations but they seem that they are not close to each other.
Title: Re: electrostatic difference potential
Post by: Anders Blom on December 1, 2010, 13:10
I believe you have attached the wrong file??

PS: don't forget the extension (.py) when attaching scripts, it makes it easier to download and open them.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 13:49
sorry I here attach the script file with extension and the log  file along with the text version of the log file.
Title: Re: electrostatic difference potential
Post by: Anders Blom on December 1, 2010, 13:52
Yes, but this is not the script you have used for the calculation, this is the Custom builder used to build the geometry.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 14:01
the script file is attached here
Title: Re: electrostatic difference potential
Post by: Anders Blom on December 1, 2010, 14:03
The system is rather large so it's probably clear that you have run out of memory (next version of ATK will use less memory!). But why do you repeat the structure 5 times along Z? Start small, with a simple structure, without repetition.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 14:09
So i should reduce the repetition...Is there any condition that only these many no.of atoms can be simulated on a pc with
1.core 2duo 1.54G
2.3GB RAM
3.256MB Graphics Card
Title: Re: electrostatic difference potential
Post by: Anders Blom on December 1, 2010, 14:11
It depends, but if you use the default double-zeta polarized basis set, with this many atoms, I would imagine you need at least 4 Gb of RAM. But the repetition doesn't add any benefit unless you want to start introducing defects etc.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 14:14
my main theme is to calculate the interaction energy in DWNT case .so is it  ok if i proceed with just 1 layer of outer and inner nanotube so that i obtain the energy values of inner,outer and combined one.
Title: Re: electrostatic difference potential
Post by: Anders Blom on December 1, 2010, 14:17
Yes, for sure.
Title: Re: electrostatic difference potential
Post by: Zexter on December 1, 2010, 14:17
ok i will try and reply you thank you