Timing-Driven Global Placement by Efficient Critical Path Extraction
By: Yunqi Shi , Siyuan Xu , Shixiong Kai and more
Potential Business Impact:
Makes computer chips faster by arranging parts better.
Timing optimization during the global placement of integrated circuits has been a significant focus for decades, yet it remains a complex, unresolved issue. Recent analytical methods typically use pin-level timing information to adjust net weights, which is fast and simple but neglects the path-based nature of the timing graph. The existing path-based methods, however, cannot balance the accuracy and efficiency due to the exponential growth of number of critical paths. In this work, we propose a GPU-accelerated timing-driven global placement framework, integrating accurate path-level information into the efficient DREAMPlace infrastructure. It optimizes the fine-grained pin-to-pin attraction objective and is facilitated by efficient critical path extraction. We also design a quadratic distance loss function specifically to align with the RC timing model. Experimental results demonstrate that our method significantly outperforms the current leading timing-driven placers, achieving an average improvement of 40.5% in total negative slack (TNS) and 8.3% in worst negative slack (WNS), as well as an improvement in half-perimeter wirelength (HPWL).
Similar Papers
Critical Path Aware Timing-Driven Global Placement for Large-Scale Heterogeneous FPGAs
Hardware Architecture
Makes computer chips run faster and more reliably.
Automated Routing-Informed Placement for Large-Scale Photonic Integrated Circuits
Emerging Technologies
Automates chip design, saving weeks of work.
Accelerating Electrostatics-based Global Placement with Enhanced FFT Computation
Hardware Architecture
Makes computer chips design much faster.