Skip to main content
more options

NGA

NGA is large-eddy simulation (LES) and direct numerical simulation (DNS) code capable of solving the low-Mach number variable density Navier-Stokes equations on structured cartesian and cylindrical meshes. It features a range of models and methods designed for solving multiphase reacting turbulent flows in complex geometries.

Additional features include:

  • Schemes that conserve mass, momentum, and kinetic energy discretely
  • Arbitrarily high order of accuracy (chosen by the user at runtime)
  • Massively parallel, with excellent scaling on 50,000 cores
  • Capability of handling massive meshes with billions of cells

Development

Stanford University

Stanford University Campus.

The development of NGA began in 2005 at the Center for Turbulence Research at Stanford University. NGA is developed jointly by Olivier Desjardins (Cornell University), Guillaume Blanquart (CalTech), and their students.

Scalability

NGA Scale-Up

NGA scale-up on NICS Kraken: NGA exhibits excellent scale-up characteristics. The largest run so far consisted of over 1.6 billion cells, and almost 50,000 compute cores.

Multiphysics Models

NGA Modules

Multiphysics modular structure of NGA: NGA consists of a number of modules capable of dealing with combustion, complex geometries, particle-laden flows, and multiphase flows.

Information Request Form

NGA licensing is managed through an inter-university agreement between Stanford, Caltech, and Cornell. To request additional information on how to obtain NGA, please fill out the form located here.