Congestion minimization during placement

Typical placement objectives involve reducing net-cut cost or minimizing wi relength. Congestion minimization is the least understood, however, it mode ls routability most accurately, In this paper, we study the congestion mini mization problem during placement. First, we show that a global placement w ith minimum wirelength has minimum total congestion. We show that minimizing wirelength may land in general, will) create locall y congested regions. We test seven different congestion minimization object ives. We also propose a post processing stage to minimize congestion. Our m ain contribution and results can be summarized as follows. 1) Among a variety of cost functions and methods for congestion minimizatio n (including several currently used in industry), wirelength alone followed by a post processing congestion minimization works the best and is one of the fastest. 2) Cost functions such as a hybrid length plus congestion (commonly believe d to be very effective) do not always work very well. 3) Net-centric post-processing techniques are among the best congestion all eviation approaches. 4) Congestion at the global placement level, correlates well with congestio n of detailed placement.