OPC Tile Scheduling for Computational Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing complexity and miniaturization of integrated circuit designs pose significant challenges for faithfully reproducing layout designs onto substrates during photolithographic manufacturing, leading to optical proximity effects and inefficiencies in resolution enhancement techniques like OPC.
Innovation Solution
The implementation of advanced scheduling techniques for resolution enhancement techniques, such as OPC, which involve dividing the layout design into regions or tiles, estimating processing times based on tile complexity and neighboring tiles' processing times, and dynamically scheduling these tasks to optimize core utilization and reduce computational waste.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If resolution enhancement techniques like OPC are applied to all tiles, then manufacturing precision is improved, but computational time and resource consumption increase significantly
Solution Approach 1:
The layout design is divided into multiple tiles, and each tile is independently analyzed and processed. This segmentation allows the system to apply resolution enhancement techniques selectively to different regions based on their individual complexity and importance, rather than uniformly processing the entire layout, thereby reducing overall computational time while maintaining precision where needed.
Solution Approach 2:
Different tiles are processed with different levels of computational intensity based on their specific characteristics. Tiles with higher complexity or critical features receive more intensive OPC processing, while simpler tiles are processed with reduced computational resources. This local differentiation optimizes the balance between manufacturing precision and computational time consumption.
2Manufacturing precision
If all tiles are processed with high computational resources, then manufacturing precision is improved, but device complexity and processing overhead increase
Solution Approach 1:
The processing system dynamically adjusts the level of computational resources allocated to each tile based on real-time analysis of tile characteristics. The system can adaptively select processing intensity, algorithm complexity, and resource allocation for each tile, allowing the overall system to maintain high precision for critical areas while reducing complexity and resource consumption for less critical regions.
3Ease of operation
If uniform processing is applied to all tiles, then ease of operation is maintained, but productivity decreases due to computational waste
Solution Approach 1:
The system automatically analyzes tile characteristics and independently determines the appropriate processing level for each tile without requiring manual intervention or complex user configuration. The automated analysis and scheduling system handles the complexity of optimizing processing resources, maintaining ease of operation while significantly improving productivity by eliminating computational waste from uniform processing of all tiles.
Data Source
AI summary
A system and method for scheduling optical proximity correction (OPC) or other resolution enhancement technique (RET) operations on a layout design is disclosed. A layout design is divided into a plurality of regions, such as a plurality of tiles. OPC is performed on the plurality of tiles in order to generate a modified layout design. Performing OPC on the plurality of tiles may be time consuming. In order to more efficiently distribute the processing of OPC, estimates of OPC processing times for the plurality of tiles is performed. The estimate of the OPC processing time for a respective tile may be based on one or both of analysis of: analysis of the respective tile; or analysis of tile(s) that neighbor the respective tile. Based on the estimates, tiles that have a longer estimated processing tile are scheduled before tiles that have a shorter estimated processing time, thereby potentially resulting in more efficient OPC processing.