You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 16 Next »

Use of the codes installed on the CNF cluster is on a self-support basis.

For training on navigating UNIX/LINUX and on using the batch system for the cluster, please contact CNF Computing staff.

Overview:

The Nanolab cluster provides users of the CNF the opportunity to use a wide range of modeling software tailored for nanoscale systems. Several classes of nodes, all linked via Gigabit Ethernet, are available on the cluster. The cluster runs Scientific Linux 7 with OpenHPC. Slurm is the batch job queuing system.

Potential users of the cluster will apply for access via the normal CNF user and project application process. 

Cluster charges are based on accessing the cluster and running jobs on the cluster – a single monthly "all you can eat" fee is charged for each month during which the cluster is used.

Head Node:

The head node is a Dell PowerEdge R740xd with approximately 9 TeraBytes of shared disk space available to users. Via NFS, user home directories are shared out to all of the compute nodes.

dn Compute Nodes:

This class comprises two Dell PowerEdge R640 systems. Each node has 256GB of memory and 2 Intel Xeon Gold 6136 processors. Each individual processor has 12 cores and 2 threads per core. This equals 48 virtual CPUs per node.

rb2u Compute Nodes:

At present, three of the twelve nodes in this class are online. We are working to provision the rest. Each node has 32GB of memory and 2 Intel Xeon E5-2620 processors. Each individual processor has 6 cores and 2 threads per core. This equals 24 virtual CPUs per node.

rb1u Compute Nodes:

None of these nodes are yet available. We are working to provision them once a new network switch is installed.

Scientific Codes on the Cluster:

MPI Families

OpenMPIhttp://www.open-mpi.orgThe Open MPI Project is an open source  Message Passing Interface  implementation that is developed and maintained by a consortium of academic, research, and industry partners.
MPICHhttp://www.mpich.orgMPICH is a high-performance and widely portable implementation of the Message Passing Interface (MPI) standard (MPI-1, MPI-2 and MPI-3). 

Compiler Families

GNUhttp://gcc.gnu.orgGNU C compiler and support files.
LLVMhttp://llvm.orgLLVM compiler infrastructure.

Python

SciPyhttp://www.scipy.orgSciPy is a Python-based ecosystem of open-source software for mathematics, science, and engineering
NumPyhttp://www.numpy.orgBase N-dimensional array package

IO Libraries

Adioshttp://www.olcf.ornl.gov/center-projects/adios

The Adaptable IO System (ADIOS).

HDF5/PHDF5http://www.hdfgroup.org/HDF5

A general purpose library and file format for storing scientific data.

NetCDFhttp://www.unidata.ucar.edu/software/netcdfFortran, C, and C++ libraries for the Unidata network Common Data
PnetCDFhttp://cucis.ece.northwestern.edu/projects/PnetCDF

A Parallel NetCDF library

ScionLibhttp://www.fz-juelich.de/ias/jsc/EN/Expertise/Support/Software/SIONlib/_node.html

Scalable I/O Library for Parallel Access to Task-Local Files.

Serial/Threaded Libraries

Rhttp://www.r-project.org

R is a language and environment for statistical computing and graphics

GSLhttp://www.gnu.org/software/gslThe library provides a wide range of mathematical routines such as random number generators, special functions and least-squares fitting. There are over 1000 functions in total.
Metishttp://glaros.dtc.umn.edu/gkhome/metis/metis/overview

Serial Graph Partitioning and Fill-reducing Matrix Ordering.

OpenBLAShttp://www.openblas.net

An optimized BLAS library based on GotoBLAS2.

Plasmahttps://bitbucket.org/icl/plasma

Parallel Linear Algebra Software for Multicore Architectures.

Scotchhttp://www.labri.fr/perso/pelegrin/scotch

Graph, mesh and hypergraph partitioning library.

SuperLUhttp://crd.lbl.gov/~xiaoye/SuperLU

A general purpose library for the direct solution of linear equations.

Parallel Libraries

Boosthttp://www.boost.org

Free peer-reviewed portable C++ source libraries.

FFTWhttp://www.fftw.org

Fast Fourier Transform library.

Hyprehttp://www.llnl.gov/casc/hypre

Scalable algorithms for solving linear systems of equations.

MFEMhttp://mfem.org

Lightweight, general, scalable C++ library for finite element methods.

MUMPShttp://mumps.enseeiht.fr

A MUltifrontal Massively Parallel Sparse direct Solver.

PETSChttp://www.mcs.anl.gov/petsc

Portable Extensible Toolkit for Scientific Computation.

PT-SCOTCHhttp://www.labri.fr/perso/pelegrin/scotch

Graph, mesh and hypergraph partitioning library using MPI.

SLEPC http://slepc.upv.esA library for solving large scale sparse eigenvalue problems.
SuperLUhttp://crd.lbl.gov/~xiaoye/SuperLUA general purpose library for the direct solution of linear equations.
Trilinos https://trilinos.org/An object-oriented software framework for the solution of large-scale, complex multi-physics engineering and scientific problems. 

Density Functional Approaches

ABINITGNU7/OpenMPIhttps://www.abinit.org/

ABINIT is a software suite to calculate the optical, mechanical, vibrational, and other observable properties of materials. Starting from the quantum equations of density functional theory, you can build up to advanced applications with perturbation theories based on DFT, and many-body Green's functions (GW and DMFT) .
ABINIT can calculate molecules, nanostructures and solids with any chemical composition, and comes with several complete and robust tables of atomic potentials.

module load lapack
module load blas
module load abinit

Linear Algebra Operations

LAPACK http://www.netlib.org/lapack/LAPACK is written in Fortran 90 and provides routines for solving systems of simultaneous linear equations, least-squares solutions of linear systems of equations, eigenvalue problems, and singular value problems. 
module load lapack
SCALAPACKGNU7/OpenMPIhttp://www.netlib.org/scalapack/

ScaLAPACK is a library of high-performance linear algebra routines for parallel distributed memory machines. ScaLAPACK solves dense and banded linear systems, least squares problems, eigenvalue problems, and singular value problems.

module load scalapack

 

Matrix and Vector Operations

BLAShttp://netlib.org/blas/The BLAS (Basic Linear Algebra Subprograms) are routines that provide standard building blocks for performing basic vector and matrix operations.
(OpenBLAS is also available... see above)
module load blas

 


Cluster HowTos:

  • Login to the Cluster
  • Loading Environments for Using Codes
  • Submitting Jobs to the Cluster
  • No labels