ComPLx: A competitive primal-dual Lagrange optimization for global placement

Abstract

We develop a projected-subgradient primal-dual Lagrange optimization for global placement, that can be instantiated with a variety of interconnect models. It decomposes the original non-convex problem into“more convex”sub-problems. It generalizes the recent SimPL, SimPLR and Ripple algorithms and extends them. Empirically, ComPLx outperforms all published placers in runtime and performance on ISPD 2005 and 2006 benchmarks.