Messages

5371

General Questions and Answers / Re: importing VNL files into VNL on Linux

« on: January 20, 2009, 20:03 »

In VNL 2008.10 there is no longer any "Lab Floor" as in old releases. The new method for importing VNL files is either to drop them directly to instruments like the Nanoscope, or use the "Result Browser". If you open this instrument (double-click the icon in the main window), you will find an "Open" button, which in addition to letting you import VNL files also supports import of xyz files.

Once the file is open in the Result Browser, you can inspect the content of the file, and drop (from the left panel) the relevant configurations to other instruments, like again the Nanoscope.

5372

General Questions and Answers / Re: Please give me advice--Charge missing problem.

« on: January 20, 2009, 11:38 »

This is a typical problem, which does not have a universal solution. There are several points to consider:

• Electrode setup
• K-point sampling
• Contour integral parameters
• Temperature

Some of this is quite advanced, but at the same time very fundamental, and perhaps not so well documented. To help users understand these points better I have therefore published a new tutorial on our web site that describes a lot of convergence tricks. Go to quantumwise.com and click on Publications>Tutorials, and you will now find there a PDF document called FeMgO Convergence Tricks.pdf.

Although I have used a spin-polarized system (FeMgO MTJ) as an example, the tips and tricks as such apply generally.

5373

Installation and License Questions / Re: Error: Terminal Server remote client not allowed.

« on: January 20, 2009, 11:25 »

This is related to the statement in our system requirements (for Windows):

"It is not possible to run a terminal-based ATK/VNL session on a remote network connection."

Basically, what it means, is that you must run VNL (and ATK) on the local machine where you sit (on Windows).

This does not, however, mean that the license server must run on the local machine, and that's the whole point of a floating license. It is even possible to tunnel connections to license servers, so if you have a license server running at the university, you can set up an SSH tunnel for the specific port specified in the license, and check out licenses even at home. This, of course, requires that you are allowed to SSH tunnel into the university network, i.e. that the firewalls are open for that kind of traffic.

The error referred to in the previous post by Serhan is also interesting, although it's related to something else, namely changing user names using the "Switch User" feature in Windows XP. I'm not sure how that would appear or affect running ATK processes, but apparently it will raise a similar error in FLEXlm with the same error code, so it's good that we now know two potential reasons for this error message.

5374

Installation and License Questions / Re: Best and fast configuration for running ATK

« on: January 19, 2009, 11:08 »

There is no doubt that ATK performs best in 64-bit on a 64-bit machine (i.e. on the x86_64 architecture). The benefit over 32-bit ATK on a 32-bit machine is about 2 times (twice as fast), and 30-50% faster compared to 32-bit ATK on a x86_64 machine.

5375

Scripts, Tutorials and Applications / Re: Script for calculating the voltage drop.

« on: January 14, 2009, 14:45 »

I think there is a fundamental misconception here, which we should clear out first.

What we mean by voltage drop is the difference in real space effective potential V(r,bias) between two calculations, i.e.

V(r,bias1) - V(r,bias0)

where the reference calculation "bias0" is for zero bias, hence the notation. The obtained quantity shows how the applied bias is distributed across the structure, i.e. where the voltage drops, i.e. where the resistance is.

The calculations posted by anyipeng (lih2li01 and lih2li-0.1) are both calculations with 0.1 V bias applied, but in two different directions. Comparing them makes no real sense, at least not in terms of "voltage drop". What you want is one calculation with no bias, i.e.

Code

electrode_voltages = (0.0,0.0)*Volt

and one calculation with bias (one of those you already made), and take the difference between these.

However, the point about using "DensityMatrix" is still very valid; one can not expect to get a proper picture of the voltage drop using the other constraints.

5376

General Questions and Answers / Re: start and end time of ATK job

« on: January 13, 2009, 22:17 »

I have activated the "Karma" feature on the Forum, which is what you asked for, Serhan So now you can "applaud" me (or anyone else for that matter) if you wish!

I called it "Reputation" however instead of Karma, which didn't seem to be an entirely suitable label. Any better suggestions? "Pythonity"? Applaud/Smite ok?

5377

General Questions and Answers / Re: vnl file is larger, how to read it?

« on: January 13, 2009, 14:26 »

There are a couple of solutions I can propose to this troublesome situation.

1) You can export the 3D grid to a file which can be imported and read by another software, like Matlab. Detailed instructions are available in another post on this forum.

2) Perhaps you can consider if you really need to visualize the voltage drop in 3D? Perhaps the real information you are looking for can be obtained by integrating the data in the XY plane and just plot how the voltage drops along the Z axis?

5378

General Questions and Answers / Re: vnl file is larger, how to read it?

« on: January 13, 2009, 14:17 »

I'm not sure the memory issue is different in Linux... Unfortunately it seems this is a real bug in VNL

5379

Installation and License Questions / Re: how to install VNL 2008.10 tried on linux

« on: January 13, 2009, 14:15 »

I'm not exactly sure what kind of license you have... The license normally depends on the MAC address (unless it's a DEMO license), but it does not depend on whether you run Windows/Linux. To run a license server requires a floating license, indicated by the word "SERVER" on the first line in the license file. Getting ATK to run in parallel is described in detail in the manual and is in principle straightforward but does require some administration skills.

5380

General Questions and Answers / MOVED: how to install VNL 2008.10 tried on linux(foroda6)

« on: January 13, 2009, 14:10 »

This topic has been moved to Installation Questions since it deals with installation issues.

http://quantumwise.com/forum/index.php?topic=67.0

5381

Installation and License Questions / Re: how to install VNL 2008.10 tried on linux

« on: January 13, 2009, 14:00 »

As quickly and simply as possible:

1. Download the package and save it in the directory under which you want to install the program (say, /opt).
2. Open a terminal and go to this directory (like, cd /opt).
3. Type tar xvfj vnl-2008.10.0-Linux-x86_64.tar.bz2 (assuming you want the 64-bit package).

That's it! (Plus installing the license, but that's a separate thing.)

5382

General Questions and Answers / Re: vnl file is larger, how to read it?

« on: January 13, 2009, 13:51 »

VNL can use a lot of memory sometimes... One way to minimize the memory usage is to not use the Result Browser, but drag the VNL file directly onto the Nanoscope, directly after starting VNL up. Does that help for your system?

(I hope you are using VNL 2008.02 or 10? Earlier versions had memory leaks...)

5383

General Questions and Answers / Re: About the parallel calculation using ATK

« on: January 13, 2009, 11:26 »

hhspace, please contact QuantumWise offline to resolve this issue. Just send me an email through my user profile here at the Forum, and we take it from there. We will post the solution once we've figured it out

For the record, however, one needs to keep two things in mind.

1) First, ATK parallelizes in two ways, or two levels. What you can see from the system monitor is the OpenMP parallelization, which shows how ATK threads on all the cores of your processor. It is not an indication, however, of the MPI parallelization we are talking about here. This is the second level of parallelization, which is over multiple network nodes.

2) The "nodes" are virtual and not related to the processor or network host. For a proper parallel job there always only a single master node (node id=0), otherwise the parallelization algorithms simply do not work properly, since the master node is responsible for collecting the data computed by the slave nodes.

It is possible to run in parallel using MPI also with a node-locked license (if you have the ATKslave feature in the license file), and it give quite ok performance on a multi-core if the job isn't too large. You will, however, in that case see several ATK processes in the system monitor.

5384

General Questions and Answers / Re: start and end time of ATK job

« on: January 13, 2009, 11:12 »

There are several options.

A rough way (which however doesn't work if you are in Windows, or if you use a queuing system to launch ATK) is to issue the ATK command wrapped in shell "date" statements, that is

Code

date ; atk myinputfile.py > logfile.log ; date 

However, this will not print the date to the logfile (logfile.log), so we're much better off using Python (as usually is the case ).

A quite fancy method, which also can calculate how long time the calculation took, is discussed in the manual.

The most convenient solution is, however, probably to just add these lines in the beginning and end of the input script:

Code

import datetime
print "Calculation started:", datetime.datetime.now().replace(microsecond=0)
...
# ... original script ...
...
print "Calculation ended:", datetime.datetime.now().replace(microsecond=0)

5385

Scripts, Tutorials and Applications / Re: Script for creating nanotubes

« on: January 12, 2009, 22:28 »

For example, let us compute the band structure of a (4,4) B-N nanotube. Now, remember that a (4,4) carbon nanotube is metallic; it turns out, the B-N is not.

Step 1: Run this script to create the geometry

Code

from ATK.KohnSham import *
from Nanotube import *

aCC = 1.45*Ang    # literature values vary from 1.43 to 1.45 Ang
n = 4
m = 4
element1 = Boron
element2 = Nitrogen

nanotube = createNanotubeAsBulk (n,m,1,(element1,element2),aCC)

vnl_file=VNLFile("bn_44.vnl")
vnl_file.addToSample(nanotube,"B-N (4,4) nanotube")

Step 2: Drop the created file on the NanoLanguage Scripter in VNL and modify the following parameters:

• Brillouin zone integration parameters: (1,1,50) is a reasonable k-point sampling
• Specify a VNL file name where to store the band structure (for instance, the same VNL file as the geometry is in). Remember to enter the full absolute file name, with path!
• Add Energy Bands to the analysis options. Delete all k-points, by right-clicking them, except the two first. Thus, we will obtain the band structure from Gamma to Z=(0,0,0.5).

Step 3: Drop the script from the NanoLanguage Scripter to the Job Manager to run it! It should take 5-10 minutes depending on your hardware.

Step 4: When the calculation finishes, drop the VNL file with the band structure on the Result Browser. Zooming in a bit, you should be able to indeed see the band gap for the (4,4) B-N nanotube as in the attached plot!