Q-Tree: a new iterative improvement approach for buffered interconnect optimization

The "chicken-egg" dilemma between VLSI interconnect timing optimization and delay calculation suggests an iterative approach. We separate interconnect timing transformation as Hanan grafting and non-Hanan sliding, and reveal generally negligible contribution of non-Hanan sliding. We propose a greedy iterative interconnect timing optimization algorithm called Q-Tree. Our experimental results show that Q-Tree starting with Steiner minimum tree topologies achieves better timing performance than C-Tree, PER-Steiner and BA-Tree algorithms. Also, executing Q-Tree starting with BA-Tree or P-Tree topologies can achieve better timing performance, especially, with shorter wires and fewer buffers. In general, Q-Tree can be applied to any interconnect tree for further timing performance improvement, with practical instance sizes and easily-extended functionality - e.g., with buffer station and routing obstacle avoidance consideration.