Due to the variety of projects we work on, different codes are utilized, depending on the problem under investigation. People in our collaboration have worked on the development of the following three codes:

• ART
• Enzo
• FLASH (Cosmology Home Page)
• HOT
• MC²

The Adaptive Refinement Tree (ART) N-body + gasdynamics code developed by Kravtsov, Klypin, and Hoffman uses a combination of particle-mesh and shock-capturing Eulerian methods for simulating the evolution of collisionless dark matter/stars and gas, respectively. The ART code uses the adaptive mesh refinement (AMR) technique to increase the resolution in the regions of interest. The AMR is particularly attractive for cosmology because the regions where the highest resolution is needed usually occupy only a small fraction of the computational volume and thus can be refined with relatively small number of mesh cells.

Enzo Enzo is an adaptive mesh refinement (AMR), grid-based hybrid code (hydro + N-Body) which was originally written by Greg Bryan and Michael Norman at the National Center for Supercomputing Applications, and is now updated and maintained by the Laboratory for Computational Astrophysics at UC San Diego. The code was originally designed to do simulations of cosmological structure formation, but has been modified to examine turbulence, galactic star formation, and other topics of interest. The code couples an adaptive particle-mesh method for solving the equations of dark matter dynamics with a hydro solver the piecewise parabolic method (PPM), which has been modified for cold, hypersonic astrophysical flows by the addition of a dual- energy formalism. In addition, the code has physics packages for radiative cooling, a metagalactic ultraviolet background, star formation and feedback, primordial gas chemistry, and turbulent driving. Enzo is freely available from the Laboratory for Computational Astrophysics.

FLASH is an Adaptive Mesh Refinement (AMR) code for treating astrophysical hydrodynamics problems. It was originally developed at the DOE ASCI Alliances Center for Astrophysical Thermonuclear Flashes at the University of Chicago for the purpose of simulating Type Ia supernovae, novae, and X-ray bursts. It has since evolved to handle more general astrophysical problems, including those involving collisionless particle dynamics. FLASH is freely available from the ASCI Flash Center.

HOT is a dark matter tree-code. This code has defined the state of the art in high-resolution cosmological N-body simulations over the last decade. An SPH hydro capability for HOT has been developed and preliminary tests have been conducted successfully. The basic algorithm underlying the HOT code may be divided into several stages. First, particles are domain-decomposed into spatial groups. Second, a distributed tree data structure is constructed. In the main stage of the algorithm, this tree is traversed independently in each processor, with requests for non-local data being generated as needed. HOT has been run on a variety of parallel platforms over the last decade and has garnered Gordon Bell awards in 1992 and 1997 (First place, performance), and 1997 (First place, price-performance).

MC² is a parallel particle mesh (PM) dark matter code, which incorporates a simplified treatment of baryons via the Hydro-Particle-Mesh (HPM) method as well as neutrino modules. The code is designed to provide excellent performance for maximum values of N_p=N_g=2048³; this number will increase as available computational resources continue to improve in size and performance. Because of the ubiquity of periodic boundary conditions in cosmology problems, MC² uses a FFT-based solver for the Poisson equation; time-stepping is handled via a symplectic method.

We have access to different machines at several institutions. These resources include:

• Livermore
  • Frost
  • Blue Pacific
  
  
• Los Alamos
  • Coyote
  • QSC
  • Pink
  • The Space-Simulator
  • Venom and Nightshade
  
  
• NCSA
  • Tungsten
  • Mercury
  • Copper
  
  
• NERSC
  • Seaborg

Currently, different data sets are available for our project:

• The Sloan Digital Sky Survey SDSS
• The Blanco Cosmology Survey BCS (Cosmology Home Page)