Topics

76

General Questions and Answers / Setting parallel in Job manager?

« on: November 3, 2015, 10:29 »

I have a license for 4 nodes (1 master, 3 slave). However the system is a local one (2xCPU - 12 cores each).

 Is it possible to set from the job manager by default that all my calculations use the full number of nodes.

In short in-stead of doing 'mpiexec -n 4 ...' each time , can I config somehow the job manager to do it for me?

77

General Questions and Answers / value (n-> inf, bulk)

« on: October 31, 2015, 17:29 »

How is value of band gap for layered material for n-> inf and bulk different from each other? 

78

General Questions and Answers / Charge density

« on: October 31, 2015, 17:25 »

How can I visualize charge density on atoms as in http://arxiv.org/pdf/1506.04707.pdf ?

79

General Questions and Answers / NDR related problem in device calculations

« on: October 31, 2015, 17:23 »

In device calculations in ATK, usually scattering free transport is considered, and for devices it often gives NDR behavior.

However, the inclusion of phonons restores the conventional shape of the IV-curves (Mathieu Luisier talk at IEDM 2014 talk on MoS2 FET: 10.1109/IEDM.2014.7047142).

His work was based on eTB with multiple neighbours, the hamiltonian being constructed by Max.localizedWannier functions. This seems a promising approach, with great accuracy and flexibility too.

Can something similar not be implemented in ATK???

80

General Questions and Answers / e-ph scattering in transport

« on: October 31, 2015, 17:18 »

Is it possible to implement e-phonon scattering in ATK transmission calculations.
Also in order to reduce phonon computation requirements, can I use a Slater-Koster approach for the phonon part only? 

81

General Questions and Answers / MAC id

« on: October 31, 2015, 17:08 »

If my system has multiple network adapters e.g. ethernet and wlan which MAC ID should be used for licensing?

82

General Questions and Answers / mobility calculation fail

« on: October 31, 2015, 15:23 »

I am trying to calculate carrier mobility in black Phosphorous, with atk 2015.rc2 , I am getting error in the electron-phonon coupling analyzer and the other results show
# Item:  0
# File:  C:/Users/sengupta/blackP_mobility.nc
# Title: blackP_mobility.nc - gID002
# Type:  HamiltonianDerivatives
Not implemented.
# Item:  0
# File:  C:/Users/sengupta/blackP_mobility.nc
# Title: blackP_mobility.nc - gID004
# Type:  Mobility
+------------------------------------------------------------------------------+
| Mobility Report                                                              |
| ---------------------------------------------------------------------------- |
| Temperature = 300.00 K                                                       |
|                                                                              |
| Hole mobility         =      -0.00 cm^2/(V*s)                                |
| Electron mobility     =      -0.00 cm^2/(V*s)                                |
|                                                                              |
| Hole conductivity     =      -0.00 S/m                                       |
| Electron conductivity =      -0.00 S/m                                       |
+------------------------------------------------------------------------------+


What is the problem?  Script attached.

83

General Questions and Answers / How many nodes

« on: September 30, 2015, 15:04 »

In http://quantumwise.com/documents/tutorials/latest/ParallelGuide/index.html/chap.quick.html#sect1.quick.nodes
I found:
--------------
For regular calculations, which are not critical in memory usage, assign one MPI process per socket (essentially a socket is a CPU, although this term is outdated).

For large calculations, which would use more than half of the RAM of each node in serial, assign one MPI process per node (that is, per physical machine).

For very small calculations, in particular with many k-points, assign one MPI process per core.
--------------

I have a 2 Intel xeon processors in my system (each have 12 cores), I have 16 GB (2x8GB) of RAM. For running optimization of large suprecells with DFT, what is the safe number of nodes I can use.

I have license for 1 master and 3 slave. So technically I can run 4 nodes?? or per CPU core (i.e. 24 nodes?)

84

General Questions and Answers / Black phophorous TB parameters

« on: September 18, 2015, 14:08 »

Can the TB parameters for black phosphorous published in http://arxiv.org/pdf/1404.0618v2.pdf be implemented in ATK?

85

Future Releases / 2015.rc2 job manager

« on: September 17, 2015, 09:03 »

How to see the log file as the calculation is running in the job manager for the 2015.rc2 ? The job manager interface seems rather complicated and it's really annoying being unable to see the log by default as a job is launched. Please correct the issue, as an advanced user it is always reassuring to see the calculation progressing real time.

86

General Questions and Answers / Saving data from script

« on: August 6, 2015, 14:05 »

I have a python script to see a plot in ATK, but I cant save it into a data file, here's the plot command
--------------------------------------
# Plot the data
import pylab
pylab.plot(energ,alpha,'b', label='plot')
pylab.xlabel(r"Energy (eV)", size=16)
pylab.ylabel(r"alpha",size=16)
pylab.show()
-------------------------

I want to save the data for plotting into a .dat file , consisting of two columns .. how do I do that?

87

General Questions and Answers / Broadening value of optical spectra

« on: August 6, 2015, 11:58 »

What should be the ideal value of the Broadening parameter in optical spectra calculator? How this would vary for say 2D and bulk materials?

88

General Questions and Answers / Incorporation of e-h interaction in optical spectra

« on: July 27, 2015, 15:30 »

Is it possible to incorporate e-h interaction in optical spectra in ATK? As Kubo-Greenwood is a single particle method can some approximate correction factor be included to consider the excitonic states?

89

General Questions and Answers / How much vacuum neccesary

« on: July 20, 2015, 16:01 »

Is a 10 Angstrom vacuum on both sides (top and bottom each) sufficeint to simulate multi-layer MoS2 (or other similar structures) if a k-point setting of 9x9x2is employed?

90

General Questions and Answers / Optical spectra e-h interaction

« on: July 10, 2015, 08:26 »

I understand from the manual that the Kubo-Greenwood formula is employed to calculate optical spectra in ATK. Is it a single particle K-G formula or it is also inclusive of e-h interactions in optical spectra ? Please reply early as I didn't get response for my earlier post.