Automated Net Grouping and Routing for IC Design

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current IC design processes require manual user intervention for net grouping and routing, leading to inefficient and time-consuming iterations, especially in complex SoC designs, where high-quality routing results are difficult to achieve without user guidance and optimization.

Innovation Solution

An automated net grouping and routing system that uses constraint graphs to determine routing order, trunk planning, and rip-up/re-planning processes to optimize net grouping and track sharing, allowing for fully automated high-quality routing without user intervention.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual user intervention is used for net grouping and routing, then routing quality can be optimized through user guidance, but the design process becomes time-consuming and inefficient

Engineering Contradiction:
Improverouting qualityVSAvoiddesign process time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The router automatically performs net grouping and routing without requiring user intervention. The system analyzes net characteristics, generates grouping strategies, and executes routing autonomously, eliminating the time-consuming manual iteration process while maintaining high routing quality through built-in optimization algorithms

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The system performs preliminary net grouping analysis before routing by examining net characteristics such as pin directions and ordering patterns. This pre-planning phase creates optimized grouping strategies that guide the subsequent routing process, ensuring high-quality results are achieved efficiently without manual intervention during routing

Inventive Principle:
Principle #10Preliminary action

2Productivity

If automated net grouping is implemented, then the design process becomes efficient and fast, but achieving high-quality routing results becomes difficult without user guidance

Engineering Contradiction:
Improvedesign process efficiencyVSAvoidrouting quality
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The automated router incorporates feedback mechanisms that analyze routing results and iteratively improve grouping strategies. The system evaluates routing quality metrics, identifies areas for improvement, and adjusts grouping decisions accordingly, enabling high-quality routing results to be achieved automatically through continuous optimization

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts routing parameters and grouping criteria based on net characteristics and routing progress. By changing parameters such as grouping thresholds, routing constraints, and resource allocation, the automated router adapts to different design scenarios and achieves high-quality results efficiently without manual intervention

Inventive Principle:
Principle #35Parameter changes

3Manufacturing precision

If complex SoC designs are routed manually, then routing quality can be maintained, but the process requires extensive user guidance and optimization iterations

Engineering Contradiction:
Improverouting qualityVSAvoiduser guidance requirements
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The automated router segments complex SoC designs into manageable net groups based on characteristic patterns such as pin directions and ordering. By dividing the routing problem into smaller, organized groups, the system handles complexity systematically without requiring user guidance, maintaining routing quality through structured automated processing

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The router implements universal automated algorithms that handle various types of nets and routing scenarios without requiring user-specific guidance. The system performs multiple functions including net analysis, grouping, constraint generation, and routing execution, eliminating the need for extensive user optimization iterations across different design complexities

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS11126780B1Automatic net grouping and routing
Publication Date: 2021.09.21 SYNOPSYS INC
  • US11126780B1 patent drawing
  • US11126780B1 patent drawing
  • US11126780B1 patent drawing

AI summary

Techniques and systems for automatic net grouping and routing are described. Some embodiments can determine a set of net groups by automatically grouping nets that have (1) a same pin count, (2) a pin direction type that is in a predefined set of pin direction types, and (3) a pin order type that is in a predefined set of pin order types. Next, the embodiments can generate routing guidance by performing trunk planning for each net group. The embodiments can then perform detailed routing for each net in each net group by using the routing guidance.