Instance-Specific Accelerators for Minimum Covering

  • Authors:
  • Christian Plessl;Marco Platzner

  • Affiliations:
  • Computer Engineering and Networks Laboratory, Swiss Federal Institute of Technology (ETH) Zurich, 8092 Zurich, Switzerland plessl@tik.ee.ethz.ch;Computer Engineering and Networks Laboratory, Swiss Federal Institute of Technology (ETH) Zurich, 8092 Zurich, Switzerland platzner@tik.ee.ethz.ch

  • Venue:
  • The Journal of Supercomputing
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents the acceleration of minimum-cost covering problems by instance-specific hardware. First, we formulate the minimum-cost covering problem and discuss a branch & bound algorithm to solve it. Then we describe instance-specific hardware architectures that implement branch & bound in 3-valued logic and use reduction techniques similar to those found in software solvers. We further present prototypical accelerator implementations and a corresponding design tool flow. Our experiments reveal significant raw speedups up to five orders of magnitude for a set of smaller unate covering problems. Provided that hardware compilation times can be reduced, we conclude that instance-specific acceleration of hard minimum-cost covering problems will lead to substantial overall speedups.