Implementing Linda for distributed and parallel processing

  • Authors:

  • Jerrold S. Leichter;Robert A. Whiteside

  • Affiliations:

  • Department of Computer Science, Yale University, P.O. Box 2158, Yale Station, New Haven, Connecticut;Scientific Computing Division, Sandia National Laboratories, Livermore, California

  • Venue:
  • ICS '89 Proceedings of the 3rd international conference on Supercomputing
  • Year:
  • 1989

Quantified Score

Hi-index 0.00

Visualization

Abstract

In a recent paper [17], we described experiments using the VAX LINDA system. VAX LINDA allows a single application program to utilize many machines on a network simultaneously. Applications implemented on the network at Sandia National Laboratories have achieved speeds considerably greater than that of a Cray-1S.In this paper, we discuss the implementation of the VAX LINDA system. The Linda language was originally conceived as a tool for programming parallel applications on multicomputers, and in fact VAX LINDA supports such programming on multiprocessor VAXes. We have demonstrated that, for suitable applications, we are able to treat an arbitrary collection of separate machines on a network as a “virtual multicomputer.” Accomplishing this requires careful implementation. It also involves some effort to get around limitations in operating systems and network implementations which were not designed with this kind of usage in mind.1