Support Options

Submit a Support Ticket

  • Discoverability Visible
  • Join Policy Invite Only
  • Created 29 Jun 2010

NEMO5 Wiki Pages

Citing NEMO5

Please see the academic license agreement (esp. sections 1.7 and 2.5.6):

We also kindly ask that you mention NEMO5 by name in the article text.

Running NEMO5

It is strongly suggested for most NEMO5 users to run simulations by submitting jobs on the RCAC computing clusters from their nanoHUB workspaces. See “Running NEMO5 Jobs on RCAC from nanoHUB”.

Building NEMO5

Please see this document for help with building NEMO5

The current manual is here:

Again, it is strongly suggested that users run NEMO5 through the nanoHUB workspace. Users may ask questions about building on the discussion forum, but we are unable to provide extensive individual support for this.

Support Materials

These and other materials can be accessed through the ‘Resources’ tab to the left:

NEMO5 2014 Summer Hands-on session materials

Regression Tests

2012 Summer School (Some of these tutorials are outdated. They can be used as a general guide to NEMO5, but many input options and features have changed.)


Please post all questions to the discussion forum

Please provide the input deck, submit command or other execution command, output files if available.

You will automatically receive emails of forum posts. To unsubscribe, uncheck the box at the bottom of the forum page


If the simulation stops quickly and the stderr shows “PythonCache File not found”, your simulation requires one or more python sources which need to be included as part of your submit statement. You can see the required file in the stdout and it will look like this: “PythonCache: trying to get file: ‘./’”

Some python sources are available here. If you find that you need a different one, please post on the forum.

Put them in the same directory as your input file and pass them to the submit command as shown below.

submit -v rossmann -i ./all.mat -i -i nemo-r19861

Input Files

See the regression tests for input files

Running NEMO5 Jobs on RCAC from nanoHUB

NanoHUB users of NEMO5 can take advantage of the computing power of the Rosen Center for Advanced Computing (RCAC). Users can submit jobs from their nanoHUB workspaces to run on the RCAC clusters. The input files are located in the user’s nanoHUB workspace, the job is submitted and run on RCAC, and the output files are written back to the user’s nanoHUB workspace. It is not necessary to compile NEMO5 or to download the executable. The materials database and an input file are required.

NanoHUB workspace:

Materials database:

Sample input files:

1. The most recent revision of NEMO5 on RCAC (Rossmann system) for use by nanoHUB workspace is 19861. Replace ‘XXXXX’ below with this number.

2. Create a symbolic link to the all.mat materials database file

ln -s /apps/share64/debian7/nemo/rXXXXX/mat/all.mat

Regardless of where you have the all.mat file located locally, the “database” line in the input file needs to be

database = ./all.mat

3. Use the submit command to request a job to run on the RCAC machines. You can find out more information about the submit command using:

submit --help

4. A typical job submission command is shown below. This particular example is located in the workspace in /apps/share64/nemo/examples/current/public_examples/bulk_GaAs_band_structure

submit -v rossmann -i ./all.mat nemo-rXXXXX

Here, the computer system rossmann is the venue where the job will execute. The materials database (all.mat) is passed with the ‘-i’ flag. The line ends with the name of the nemo executable and the input file.

For instance, if the materials database file is in a separate directory, the command may look something like

submit -v rossmann -i ./all.mat nemo-rXXXX

After the job runs, output files will be written back to your nanoHUB workspace directory where you submitted the job.

5. Larger simulations can be run on multiple cores. See the “Partitioning” section (4.2.4) of the NEMO5 manual. The total number of cores ‘-n’ and wall time ‘-w’ can specified. For ‘-N’, the value should be equal the number of cores per node. See “RCAC Machine Constraints” section below.

The submit command takes arguments related to the number of processors and walltime required.

submit --help

The relavent arguments are

  -n NCPUS, --nCpus=NCPUS   Number of cores for MPI execution
  -N PPN, --ppn=PPN     Number of cores/node for MPI execution
  -w WALLTIME, --wallTime=WALLTIME

The following job requests 24 cores on Rossmann for 5 minutes (the example below is located in public_examples/bulk_GaAs_parallel)

submit -v rossmann -i ./all.mat -n 24 -w 0:05:00 nemo-rXXXX

This will request 16 cores on two nodes

submit -v rossmann -i ./all.mat -n 48 -N 24 -w 0:05:00 nemo-rXXXX

If the request is for more than one total node, the request must be for the total number of cores on the nodes. In other words, one cannot have

-n 36 -N 24

However, one can have

-n 1 -N 1

RCAC Machine Constraints

Job submissions to rossmann go to the standby queue, which has a maximum walltime of 4 hours.

There are currently three venues available.

Rossman has 24 cores per node and Conte has 16 cores per node.

standby@rossmann: 4 hours, maximum_cores=TBD
ncn-hub@rossmann: 720 hours, maximum_cores=TBD
ncn-hub@conte: 336 hours, maximum_cores=16

You can specify the queue (“standby@rossmann”) or leave it blank (“rossmann”) and the submit command will try to find a venue for the job

Input Deck and Materials Database Editor

For use with Firefox browser

Static Binary and Source Code

NEMO5 Citations

NEMO5 has been using the following works:

B. Bishnoi, B. Ghosh, “p+-Al0.3Ga0.7Sb Pocket-Implanted L-shapedGaSb/InAs Heterojunction Vertical n-TFETs”, arxiv, Aug. 2014.



  • Michael Povolotskyi: (atomistic) geometry generation, FEM (libmesh), Poisson solver (libmesh), quadrature rules (libmesh), degree-of-freedom map, input deck parser, general simulation class and simulation flow, Makefile system
  • Tillmann Kubis: tight-binding Hamiltonian and passivation, Propagation class
  • Jim Fonseca: memory management, strain, support, testing
  • Jean Michel Sellier: Structure design screen of 1d Heterostructure Tool
  • Zhengping Jiang: effective mass Hamiltonian, resonance finder, RTD simulations


  • Hong-Hyun Park: transport simulations (NEGF with recursive algorithms, wavefunction formalism)
  • Sebastian Steiger: material database, matrix classes, PETSc matrices and linear solvers, SLEPc eigensolvers, messaging system, K-space, QHULL tetrahedralizer and Voronoi cell generator, VFF strain solver, VFF phonon solver, most crystal classes, test suite
  • Lang Zeng: random dopants
  • Denis Areshkin: 2nd derivative of some VFF energy terms, 2D irreducible wedge finder
  • Arun Goud: Nanohub interfaces to 1d Heterostructure Tool, Quantum Dot Lab, Brillouin Zone Viewer, and Resonant Tunneling Diode Simulation using NEGF
  • Lars Bjaalie: Nanohub interface to Quantum Dot Lab
  • Benjamin Haley: material database (initial version)

Created on , Last modified on, a resource for nanoscience and nanotechnology, is supported by the National Science Foundation and other funding agencies. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.