Wire length prediction based clustering and its application in placement

  • Authors:
  • Bo Hu;Malgorzata Marek-Sadowska

  • Affiliations:
  • Univ. of California, Santa Barbara, CA;Univ. of California, Santa Barbara, CA

  • Venue:
  • Proceedings of the 40th annual Design Automation Conference
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we introduce a metric to evaluate proximity of connected elements in a netlist. Compared to connectivity [8] and edge separability [4], our metric is capable of predicting short connections more accurately. We show that the proposed metric can also predict relative wire length in multi-pin nets. We develop a fine-granularity clustering algorithm based on the new metric and embed it into the Fast Placer Implementation (FPI) framework [10]. Experimental results show that the new clustering algorithm produces better global placement results than the net absorption [10] algorithm, connectivity [8], and edge separability [4] based algorithms. With the new clustering algorithm, FPI achieves up to 50% speedup compared to the latest version of Capo8.5 [19], without placement quality losses.