Integrated Circuit Global Routing via Localized Region Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current integrated circuit design tools lack efficient global routing optimization, as they primarily focus on synthesis and placement, failing to address varying design challenges across different chip areas, such as timing, power, routability, congestion, and yield, which require tailored optimization objectives and constraints.
Innovation Solution
The method involves defining regions and netgroups within integrated circuits, associating optimization objectives with each, generating constraints based on these objectives, and performing global routing using a revised resource sharing algorithm to optimize each region or netgroup independently, allowing for fine-grained optimization across the chip.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If global routing is performed without localized optimization, then the routing process is simpler and faster, but it fails to address specific design challenges in different chip areas such as timing, power, routability, congestion, and yield
Solution Approach 1:
The chip is divided into multiple regions, each with its own optimization objectives and constraints. The routing process is segmented into region-specific optimizations for timing, power, routability, congestion, and yield, allowing tailored optimization for each area while maintaining overall routing efficiency
Solution Approach 2:
Different optimization objectives are applied to different regions of the chip based on local design requirements. Each region can have customized constraints and optimization goals, enabling localized quality improvement without requiring complete redesign of the entire routing system
2Reliability
If optimization is applied to the entire chip uniformly, then the routing process is simpler to manage, but it cannot address varying design optimization objectives across different regions
Solution Approach 1:
The optimization process is divided into region-specific segments, each handling a particular optimization objective (timing, power, routability, congestion, yield). This segmentation improves reliability by addressing each objective with appropriate constraints while managing complexity through modular organization
Solution Approach 2:
The system dynamically assigns different optimization objectives to different regions based on design requirements. The routing process can adaptively adjust constraints and objectives for each region, improving overall routing quality while maintaining manageable complexity through dynamic configuration
Data Source
AI summary
The present disclosure relates to methods, processing systems and computer program products of global routing of integrated circuits based on localized routing optimization. In certain embodiments, the method may include one or more of: defining one or more regions, one or more netgroups, and combinations thereof of an integrated circuit, associating at least one optimization objective with each region and/or each netgroup defined, generating one or more constraints for each region and/or each netgroup based on the associated optimization objectives, and performing global routing of the integrated circuit according to the one or more constraints.