Circuit Design Modification Using Similar Prior Simulations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complexity of modern integrated circuit (IC) designs necessitates improved electronic design automation (EDA) tools to optimize circuit designs efficiently and reduce computation costs, as existing methods face large runtimes and exponentially large search spaces in design parameters.
Innovation Solution
A machine learning-based approach that leverages previous circuit design sessions to identify similarities and apply simulation data to modify current designs, utilizing warmstart and coldstart sessions to refine search spaces and reduce computation requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If traditional EDA tools are used to optimize circuit designs, then design quality can be maintained, but computation time and runtime increase exponentially with design complexity
Solution Approach 1:
The system performs preliminary actions by conducting warmstart sessions that leverage simulation data from previous similar circuit designs before initiating full optimization. This pre-computation of relevant simulation data from historically similar designs reduces the search space and computation time required for current design optimization while maintaining design quality through subsequent refinement steps.
Solution Approach 2:
The system creates and utilizes copies of simulation data from previous circuit design sessions. By identifying similar historical designs and copying their simulation results and parameter values, the system avoids redundant computations and accelerates the optimization process for current designs with comparable characteristics.
2Reliability
If exhaustive search methods are used to explore design parameters, then optimal design solutions can be found, but the search space becomes exponentially large and computationally infeasible
Solution Approach 1:
The system applies local quality by focusing simulation efforts on specific, relevant portions of the design parameter space rather than performing exhaustive global searches. By identifying similar historical designs and copying their parameter configurations, the system concentrates computational resources on locally optimal regions that are most likely to yield successful results for the current design.
Solution Approach 2:
The system utilizes parameter changes by adjusting and refining design parameters based on simulation data from similar historical designs. Rather than exhaustively searching all possible parameter combinations, the system starts with parameter values copied from similar designs and iteratively refines them through targeted simulations, thereby managing search space complexity while maintaining optimization quality.
3Manufacturing precision
If simulation data from all previous design sessions is applied to current designs, then comprehensive optimization can be achieved, but data relevance decreases and computation costs increase
Solution Approach 1:
The system performs preliminary similarity assessment between current designs and historical designs before applying simulation data. This pre-filtering step identifies the most relevant historical designs, ensuring that only high-quality, relevant simulation data is copied and applied, thereby maintaining optimization quality while avoiding the computational overhead of processing all historical data.
Solution Approach 2:
The system applies local quality by selectively copying simulation data only from historically similar designs rather than applying data from all previous sessions. This targeted approach ensures high data relevance and maintains optimization quality while significantly reducing data processing complexity and computation costs compared to comprehensive data application.
Data Source
AI summary
A method of electronic design automation (EDA) using machine learning to modify a current circuit design is provided. The method includes searching for data associated with previous design sessions for previous circuit designs, the searching being performed by implementing machine learning to identify similarities between the current circuit design and the previous circuit designs and the searching providing simulation data identifying the previous design sessions and simulation results of the previous design sessions, determining, in dependence on the simulation data, a respective degree of relevance between the current circuit design and the previous circuit designs, and performing a modification by applying the simulation data to the current circuit design based on the degree of relevance, the applying of the simulation data including performing simulations that apply values obtained from the simulation data to parameters related to the current circuit design and settings related to a simulation of the current design.