An accurate detailed routing routability prediction model in placement

Abstract

Routability is one of the primary objectives in placement. There have been many researches on forecasting routing problems and improving routability in placement but no perfect solution is found. Most traditional routability-driven placers aim to improve global routing result, but true routability lies in detailed routing. Predicting detailed routing routability in placement is extremely difficult due to the complexity and uncertainty of routing. In this paper, we propose a new detailed routing routability prediction model based on supervised learning. After extracting key features in placement and detailed routing, multivariate adaptive regression is performed to train the connection between these two stages. Using a well-trained model, most design rule violations after detailed routing can be foreseen in placement stage. Experiments show that our average prediction accuracy is 79.8%, which is comparable with other state-of-art routability estimation techniques.

How Deep Learning Can Drive Physical Synthesis Towards More Predictable Legalization

“…Recently, machine learning techniques have been used to predict the violation of routability constraints in a placed netlist. For example, the work from [19] extracts features regarding pin distribution, routing blockage, global routes and local nets in order to predict the number of DRC violations in a placed area. The work from [2] improves this idea by predicting the actual locations of the DRC violations, using a different set of features.…”

“…The high flexibility provided by machine learning (ML) models allows their use to predict the outcome of physical design algorithms. They have been employed so far to help choose between different clock tree synthesis algorithms [11], to fix miscorrelations between different timing engines [7], and to identify detailed routing violations during the placement stage [2,17,19]. The benefits of ML models come from their ability to improve the quality of physical design algorithms by predicting information that would otherwise be too costly to evaluate during execution.…”

A local congestion elimination technique driven by overflow

“…Machine learning has become pervasive in various research fields and commercial applications, and achieved satisfactory products [12]. Supervised learning is employed to detect detailed routing violations [13,14,15]. A machine learning framework is proposed to predict detailed routing short violations just from a placed netlist [16].…”

RouteNet

“…To find accurate yet fast routability prediction approach, people recently explore machine learning techniques, which have exhibited exciting progress and have been investigated in several EDA applications [11], including lithography hotspot detection [7], structured design placement [23] and NoC router modeling [10]. In [18,26], Multivariate Adaptive Regression Spline (MARS) is applied for forecasting detailed routing routability. However, global routing solution is still required as an input feature to the learning here so that there is no benefit for runtime reduction.…”

“…MARS and Support Vector Machine (SVM)-based routability forecast without using global routing information is proposed in [3]. However, this technique does not indicate how to handle macros, which prevail in modern chip designs and considerably increase the difficulty of routability prediction [26]. In [4], an SVM-based method is introduced for predicting locations of DRC hotspots.…”