The Implementation of a Coherent Memory Abstraction on a NUMA (Non-Uniform Memory Access) Multiprocessor: Experiences with PLATINUM (Platform for Investigating Non-Uniform Memory). Revision

Abstract

PLATINUM is an operating system kernel with a novel memory management system for Non-uniform Memory Access (NUMA) multiprocessor architectures. This memory management system implements a coherent memory abstraction. Coherent memory is uniformly accessible from all processors in the system. When used by applications coded with appropriate programming styles it appears to be nearly as fast as local physical memory and it reduces memory contention. Coherent memory makes programming NUMA multiprocessors easier for the user while attaining a level of performance comparable with hand-tuned programs. This paper describes the design of the PLATINUM memory management system with emphasis on the implementation of coherent memory and the factors that affect its performance. We measure the cost of basic operations implementing coherent memory. We also measure the performance of a set of application programs running on PLATINUM. Finally, we comment on the interaction between architecture and the coherent memory system. Platinum currently runs on the BBN Butterfly Plys (TM) Multiprocessor. (KR)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 06, 1989
Accession Number
ADA213909

Entities

People

  • Alan L. Cox
  • Robert J. Fowler

Organizations

  • University of Rochester

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Application Software
  • Computations
  • Computer Programming
  • Computer Science
  • Computers
  • Information Systems
  • Instrumentation
  • Language
  • Lepidoptera
  • Operating Systems
  • Parallel Computing
  • Parallel Processing
  • Parallel Processors
  • Programming Languages
  • Simulations
  • Simulators

Fields of Study

  • Computer science
  • Engineering

Readers

  • Parallel and Distributed Computing.