TDR: A Distributed-Memory Parallel Routing Algorithm for FPGAs

  • Authors:
  • Lucídio A. F. Cabral;Júlio S. Aude;Nelson Maculan

  • Affiliations:
  • -;-;-

  • Venue:
  • FPL '02 Proceedings of the Reconfigurable Computing Is Going Mainstream, 12th International Conference on Field-Programmable Logic and Applications
  • Year:
  • 2002

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a distributed-memory parallel routing algorithm for FPGAs based on the partitioning of the routing graph under special FPGA architectural constraints. Coarse-grain parallelism is adopted by assigning different processors to the routing of nets within different partitions of the routing graph. The experimental results have demonstrated that TDR can achieve linear and superlinear speedups in relation to the state-of-art VPR router even when it is running on an Ethernet cluster of workstations. However, it usually requires more tracks than VPR for routing the same circuit.