Difference between revisions of "Main Page"

From Nekcem
Jump to navigationJump to search
Line 1: Line 1:
 
<big>'''Features'''</big>
 
<big>'''Features'''</big>
  
NekCEM is a high-fidelity electromagnetic solver that has been developed at Mathematics and Computer Science Division of Argonne National Laboratory. It's an open source code, written by Misun Min, Jing Fu, Andreas Kloeckner in 1996-2011, with technical inputs from Paul Fischer [http://www.mcs.anl.gov/~fischer] and his incompressible Navier-Stokes solver Nek5000[http://nek5000.mcs.anl.gov].  
+
NekCEM https://svn.mcs.anl.gov/repos/NEKCEM is a high-fidelity electromagnetic solver that has been developed at Mathematics and Computer Science Division of Argonne National Laboratory. It's an open source code, written by Misun Min, Jing Fu, Andreas Kloeckner in 1996-2011, with technical inputs from Paul Fischer [http://www.mcs.anl.gov/~fischer] and his incompressible Navier-Stokes solver Nek5000[http://nek5000.mcs.anl.gov].  
 
The code is written in Fortran and C, using MPI for parallel communication.
 
The code is written in Fortran and C, using MPI for parallel communication.
  

Revision as of 13:42, 12 June 2011

Features

NekCEM https://svn.mcs.anl.gov/repos/NEKCEM is a high-fidelity electromagnetic solver that has been developed at Mathematics and Computer Science Division of Argonne National Laboratory. It's an open source code, written by Misun Min, Jing Fu, Andreas Kloeckner in 1996-2011, with technical inputs from Paul Fischer [1] and his incompressible Navier-Stokes solver Nek5000[2]. The code is written in Fortran and C, using MPI for parallel communication.

The code targets high performance high-order simulations on the advanced computer architectures for the applications in accelerator physics and nanoscience, predicting optimal designs of next-generation electromagnetic devices such as accelerator components for the International Linear Collider or the Large Hadron Collider, nanosensors for molecular detection, and photovoltaic solar cells with high energy-conversion efficiency.


  • High-order spectral element discretizations
  • Hexahedral boody conforming meshes
  • The 4th-order Runge-Kutta timestepping
  • The high-order exponential time integration
  • high parallel efficiency scaling over 100,000 cores
  • parallel IO scaling over 65,000 cores
  • Light transmission calculations for nanodevices
  • Wakepotential calculations for accelerator devices


Upcoming

  • Hybrid programming
  • Parallel IO with pthreading


Instruction


Current Developers

Misun Min [3], Jing Fu [4]


Getting started

Consult the User's Guide for information on using the wiki software.