首页> 外文会议>IEEE VLSI-TSA International Symposium on VLSI Design >Faster min-cut computation in unweighted hypergraphs/circuit netlists
【24h】

Faster min-cut computation in unweighted hypergraphs/circuit netlists

机译:在未加权的超图/电路网手中更快地切割计算

获取原文

摘要

A hypergraph minimum cut algorithm which is the fastest algorithm to date for computing a minimum cut in a weighted hypergraph was presented in W. K. Mak and D. F. Wong (2000). It is based on the repeated application of a special minimum s-t cut procedure. Unlike most minimum s-t cut procedures, it does not rely on any flow computation. And it runs in O(p + nlogn) time for a general weighted hypergraph where p is the total number of terminals of all hyperedges and n is the number of nodes in the input hypergraph. In this paper, we reduce its running time to O(p) if the input hypergraph is unweighted. Since a circuit netlist is an unweighted hypergraph where p = O(n) typically, this improvement can reduce the running time in circuit partitioning application significantly.
机译:一种超图最小剪辑算法,其迄今为止迄今为止算法的最快算法,其介绍了加权超图中的最小切割。它基于对特殊的最小S-T切割程序的重复应用。与最低最低的S-T剪切程序不同,它不依赖于任何流量计算。并且它在O(P + NLogn)时间内运行一般加权的超图,其中P是所有超高度的终端的总数,n是输入超图中的节点的数量。在本文中,如果输入超图是未加速的,我们将其运行时间降低到O(p)。由于电路网手册是一个未加权的超图,其中P = O(n)通常,这种改进可以显着减少电路分区应用中的运行时间。

著录项

相似文献

  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号