A gentle introduction to the associative computing (ASC) model of parallel computation

  • Authors:
  • Michael Scherger

  • Affiliations:
  • Texas Christian University, Fort Worth, TX

  • Venue:
  • Journal of Computing Sciences in Colleges
  • Year:
  • 2014

Quantified Score

Hi-index 0.00

Visualization

Abstract

Recent computing hardware technologies such as the Intel Xeon Phi and NVIDIA CUDA family of hardware coprocessors have had tremendous impact on how computer scientists program parallel computers. As an alternative to the traditional parallel thread based models (OpenMP) or multicomputer message passing models (MPI), associative processing is the concept of extending the use of associative memories into a more general parallel computing environment. The Associative Computing model (ASC) is both a model of parallel computation and a programming language. Associative computing includes the use of data parallel programming and the ability to search memory by data content rather than memory address. Unlike programming MIMD computers, ASC programmers are not responsible for task allocation, load balancing, synchronization points, etc. This paper serves a gentle introduction to the ASC model. The properties of the ASC model and hardware realizations are discussed along with examples of using the ASC programming language.