Application of Advanced Multi-Core Processor Technologies to Oceanographic Research

Abstract

The long-term goal is to improve our ability to sense and predict ocean processes, utilizing state-of-the-art information processing architectures. Next-generation processor architectures (multi-core, multi-threaded) hold the promise of delivering enormous amounts of compute power in a small form factor and with low power requirements. However, new programming models are required to realize this potential. Our objectives are to deploy signal processing algorithms onto a variety of systems on a chip (SOC) such as those being developed by Intel and NVidia, as well as the application of SOC architectures for other vehicle functions. The overarching theme of this work relates to the application of advanced heterogeneous processors (both in an embedded environment and in a cluster) to high bandwidth signal processing. Our previous work included the development of a task dispatcher model for rapid development of signal processing applications on the IBM Cell/B.E. platform. This year, we completed several enhancements to this system; including the addition of scheduling tasks on a cluster, heterogeneous (GPU/CPU) computation, and a graphical programming language.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 30, 2012
Accession Number
ADA590453

Entities

People

  • Mark R. Abbott

Organizations

  • Oregon State University

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Beam Forming
  • Computations
  • Computer Programming
  • Computing System Architectures
  • Digital Signal Processing
  • Information Processing
  • Language
  • Lessons Learned
  • Parallel Computing
  • Parallel Processing
  • Processing Equipment
  • Programming Languages
  • Signal Processing
  • Software Defined Radio
  • Universities

Fields of Study

  • Computer science

Readers

  • Distributed Systems and Data Platform Development
  • Parallel and Distributed Computing.