Hierarchical Grids for IC Layout Data Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing approaches to managing and querying hierarchical electronic design data face inefficiencies in memory usage and query performance due to the need for duplicate references and excessive memory consumption, particularly when dealing with large geometric elements and fill shapes in integrated circuit layouts.
Innovation Solution
The implementation of hierarchical grids organizes physical objects into a set of gradually refined grids, eliminating the need for duplicate references and enhancing query performance by using a two-level search mechanism and logical partitioning to efficiently track and manage layout data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If hierarchical approach with instances is used to store design data, then memory usage is improved, but query performance deteriorates due to the need to traverse the entire hierarchy
Solution Approach 1:
The patent segments the hierarchical design data into a two-level structure: a top-level hierarchy for high-level design elements and a bottom-level grid for detailed geometric elements. This segmentation allows the system to maintain the memory efficiency of hierarchical storage while enabling fast queries through the grid structure, as queries can directly access the bottom level without traversing the entire hierarchy.
Solution Approach 2:
The patent introduces a bottom-level grid as an intermediary structure between the top-level hierarchy and the detailed geometric data. This intermediary grid acts as a mediator that provides direct access to geometric elements while maintaining the hierarchical organization, thus resolving the contradiction between memory efficiency and query performance.
2Productivity
If flattened design data is used to improve query efficiency, then search speed is improved, but memory consumption increases excessively
Solution Approach 1:
The patent segments the design data into hierarchical levels, storing detailed geometric elements at the bottom level while maintaining references to master designs at the top level. This segmentation enables fast searches through the grid structure while avoiding the memory consumption of fully flattened data, as only necessary geometric details are stored at the bottom level.
3Reliability
If duplicate references are maintained for all instances, then data completeness is improved, but device complexity increases
Solution Approach 1:
The patent extracts duplicate references from the data structure by using a reference-based hierarchical model. Instead of storing complete duplicate geometric data for all instances, the system stores references to master designs and only maintains geometric data at the bottom level when necessary, thus reducing data structure complexity while preserving data completeness through the hierarchical reference chain.
Data Source
AI summary
Disclosed is an improved approach for managing, tracking, and querying hierarchical data in layouts. According to some aspects, hierarchical grids are employed utilizing a scheme that organizes physical objects into a set of gradually refined grids that avoids the need to maintain duplicates while enhancing the desirable characteristics of existing schemes, including fast query times, fast data structure initialization and reduced memory footprint. Each grid-cell may be further partitioned into sub-containers to more efficiently provide space pruning during query operations. According to one approach, structures maintained to track existence of objects in a descendent hierarchy.