Corpus ID: 26529534

Parallel Programming with Matrix Distributed Processing

@article{Pierro2005ParallelPW,
  title={Parallel Programming with Matrix Distributed Processing},
  author={Massimo Di Pierro},
  journal={ArXiv},
  year={2005},
  volume={abs/hep-lat/0505005}
}
  • M. D. Pierro
  • Published 9 May 2005
  • Physics, Computer Science
  • ArXiv
Matrix Distributed Processing (MDP) is a C++ library for fast development of efficient parallel algorithms. It constitues the core of FermiQCD. MDP enables programmers to focus on algorithms, while parallelization is dealt with automatically and transparently. Here we present a brief overview of MDP and examples of applications in Computer Science (Cellular Automata), Engineering (PDE Solver) and Physics (Ising Model). 
2 Citations
Latfield2: A c++ library for classical lattice field theory
TLDR
It is a significant rewrite of the latfield framework, moving from a slab domain decomposition to a rod decomposition, where the last two dimension of the lattice are scattered into a two dimensional process grid. Expand
An object oriented code for simulating supersymmetric Yang-Mills theories
TLDR
The details of a C++ code, which can be used to simulate supersymmetric Yang–Mills theories, are described, with particular emphasis being placed on SYM theories with N = 2 in two dimensions and N = 4 in three and four dimensions, making one-to-one comparisons between the essential components of the SYM theory and their corresponding counterparts appearing in the simulation code. Expand