Research Article

On the conditions necessary for removing abstraction penalties in OOLALA

  1. Mikel Luján*,
  2. T. L. Freeman,
  3. John R. Gurd

Article first published online: 23 FEB 2005

DOI: 10.1002/cpe.857

Issue

Concurrency and Computation: Practice and Experience

Concurrency and Computation: Practice and Experience

Special Issue: 2002 ACM Java Grande–ISCOPE Conference Part II

Volume 17, Issue 7-8, pages 839–866, June - July 2005

Additional Information

How to Cite

Luján, M., Freeman, T. L. and Gurd, J. R. (2005), On the conditions necessary for removing abstraction penalties in OOLALA. Concurrency Computat.: Pract. Exper., 17: 839–866. doi: 10.1002/cpe.857

Author Information

  1. Centre for Novel Computing, University of Manchester, Oxford Road, Manchester M13 9PL, U.K.

*Centre for Novel Computing, University of Manchester, Oxford Road, Manchester M13 9PL, U.K.

Publication History

  1. Issue published online: 23 FEB 2005
  2. Article first published online: 23 FEB 2005
  3. Manuscript Accepted: 14 OCT 2003
  4. Manuscript Revised: 1 SEP 2003
  5. Manuscript Received: 15 JAN 2003

Funded by

  • Posdoctoral Fellowship from the Department of Education, Universities and Research of the Basque Government

SEARCH

SEARCH BY CITATION

ARTICLE TOOLS

Get PDF (315K)

Keywords:

  • object-oriented programming;
  • numerical linear algebra;
  • Java;
  • abstraction penalty

Abstract

OOLALA is an object-oriented linear algebra library designed to reduce the effort of software development and maintenance. In contrast with traditional (Fortran-based) libraries, it provides two high abstraction levels that significantly reduce the number of implementations necessary for particular linear algebra operations. Initial performance evaluations of a Java implementation of OOLALA show that the two high abstraction levels are not competitive with the low abstraction level of traditional libraries. These initial performance results motivate the present contribution—the characterization of a set of storage formats (data structures) and matrix properties (special features) for which implementations at the two high abstraction levels can be transformed into implementations at the low (more efficient) abstraction level. Copyright © 2005 John Wiley & Sons, Ltd.

Get PDF (315K)