Open-source GPU-accelerated VLSI Placement


We present DREAMPlace, a deep learning toolkit-enabled VLSI placement framework. With the analogy between nonlinear VLSI placement and deep learning training problem, this tool is developed with deep learning toolkit for flexibility and efficiency. The tool runs on both CPU and GPU. Over 30X speedup over the CPU implementation (RePlAce) is achieved in global placement and legalization on ISPD 2005 contest benchmarks with a Nvidia Tesla V100 GPU.

Dummy Fill Insertion


We present DFI, a fill optimization framework with multiple objectives and constraints.