SPRNG

From HP-SEE Wiki

Revision as of 13:13, 17 November 2011 by Gurov (Talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Contents


Authors/Maintainers

SPRNG is a project of the Department of Computer Science and the School of Computational Science at Florida State University. For technical assistance in installing or using SPRNG, users should send email to sprng@cs.fsu.edu.

Summary

Pseudorandom number generation algorithms are key components in many of modern scientific computing applications and the quality of the obtained research results may depend on their characteristics. The random number generating algorithms should be of sufficient quality, so as to allow research to be focused on the scientific problem itself, and not to require detailed study of the behaviour of the generator. SPRNG (Scalable Parallel Random Number Generators) is one of the best libraries for generating pseudorandom numbers for parallel applications. It provides user-friendly interfaces for parallel C/C++ and Fortran applications, which enable the user to easily obtain multiple and sufficiently uncorrelated pseudorandom number streams on different processors, with no inter-process communication. The algorithms implemented in SPRNG are: Combined Multiple Recursive Generator, the 48 bit Linear Congruential Generator, the 64 bit Linear Congruential Generator, the modified Lagged Fibonacci Generator, the multiplicative Lagged Fibbonacci Generator, and the Prime Modulus Linear Congruential Generator.

Features

  • SPRNG Generators:
    • Combined Multiple Recursive Generator
    • 48 bit Linear Congruential Generator
    • 64 bit Linear Congruential Generator
    • Modified Lagged Fibonacci Generator
    • Multiplicative Lagged Fibbonacci Generator
    • Prime Modulus Linear Congruential Generator
  • Different interfaces provided for C/C++ and Fortran.
  • A test suite with statistical tests and tests based on physical models (ex ISING model) is provided.
  • SPRNG is easy to extend, and new parallel generators can easily be developed to support SPRNG features.

Architectural/Functional Overview

  • high level design info, how it works, performance - may be a link, or several links

Usage Overview

Dependencies

  • list of all relevant dependencies on other libraries

HP-SEE Applications

  • NUQG Numerical study of ultra-cold quantum gases
  • SETSimulation of electron transport
  • GIM Geophysical Inversion Modeling

Resource Centers

Usage by Other Projects and Communities

  • If any

Installation

Recommendations for Configuration and Usage