Smart Regression Planner for Constrained Random Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Constrained random verification (CRV) faces challenges in efficiently exploring large parameter spaces, leading to sub-optimal coverage and increased costs due to the complexity of design verification, where human engineers may miss trends or draw incorrect conclusions, resulting in inefficient verification processes.
Innovation Solution
The implementation of automatic test parameter tuning using a Smart Regression Planner (SRP) framework that formulates test parameter configuration as a black-box optimization problem, employing Bayesian optimization or other methods to iteratively optimize parameters and achieve desired verification coverage, thereby reducing the need for manual intervention and improving coverage convergence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual parameter exploration is used in constrained random verification, then verification engineers can control the testing process, but the search space becomes too large to fully explore and coverage convergence becomes inefficient
Solution Approach 1:
The system implements self-service by automatically optimizing test parameters through Bayesian optimization algorithms. The verification system performs self-adjustment of parameters based on coverage feedback, eliminating the need for manual parameter tuning while improving coverage convergence speed. The system serves itself by autonomously exploring the parameter space and identifying optimal configurations.
Solution Approach 2:
The patent replaces the mechanical manual parameter exploration process with an automated computational system. Instead of engineers manually adjusting parameters, a Bayesian optimization algorithm automatically explores the parameter space, analyzes coverage results, and iteratively optimizes parameters. This substitution transforms manual mechanical adjustment into an automated intelligent system.
2Reliability
If the number of test parameters increases to handle design complexity, then verification coverage can be improved, but the search space grows exponentially making it harder for humans to reason about and optimize
Solution Approach 1:
The patent introduces an intermediary Bayesian optimization algorithm that mediates between the complex parameter space and the verification objectives. This intermediary system processes the high-dimensional parameter space, evaluates coverage outcomes, and guides parameter optimization without requiring human engineers to directly reason about the complex interactions. The algorithm acts as a bridge between parameter configuration and coverage achievement.
Solution Approach 2:
The system dynamically changes parameters based on optimization results and coverage feedback. Instead of manually setting parameters, the system automatically adjusts parameter values through iterative optimization, transforming the static manual parameter setting process into a dynamic adaptive process that navigates complex parameter spaces efficiently.
3Extent of automation
If verification engineers manually optimize parameters, then they can apply domain knowledge, but human engineers may miss trends or draw incorrect conclusions resulting in sub-optimal coverage and runtime
Solution Approach 1:
The patent implements a feedback-driven optimization loop where coverage results from each regression iteration are analyzed and fed back into the Bayesian optimization algorithm. This continuous feedback mechanism allows the system to learn from previous results, identify trends, and adjust parameters accordingly. The feedback loop replaces human analysis with automated statistical analysis that does not miss trends or draw incorrect conclusions.
Solution Approach 2:
The system performs preliminary analysis of coverage results automatically before proceeding to parameter optimization. By pre-processing and analyzing coverage data through statistical methods, the system prepares accurate insights that guide subsequent parameter adjustments, eliminating the need for human engineers to manually analyze trends and reducing the risk of incorrect conclusions.
4Reliability
If exhaustive exploration of the parameter space is attempted, then optimal coverage can be achieved, but the verification process becomes too time-consuming and costly
Solution Approach 1:
The patent applies partial action by using Bayesian optimization to explore only the most promising regions of the parameter space rather than exhaustively searching all possibilities. The algorithm identifies and focuses on parameter configurations likely to yield high coverage, performing sufficient but not excessive exploration. This approach achieves near-optimal coverage with significantly reduced verification time compared to exhaustive search.
Solution Approach 2:
The system performs preliminary parameter optimization through iterative Bayesian optimization before final verification. By pre-optimizing parameters through multiple regression iterations with automated analysis, the system converges to optimal or near-optimal parameter settings efficiently, reducing the time required for exhaustive exploration while maintaining high coverage completeness.
Data Source
AI summary
This document discloses systems and methods for implementing automatic test parameter tuning in constrained random verification. In aspects, a method receives a first set of parameters for testing a design under test, performs a first regression (e.g., an overnight regression test) on a design under test using the first set of parameters, and analyzes the results of the first regression including determining a coverage percentage. The method then generates an optimized set of parameters based on the analysis of the results of the first regression and performs an additional regression on the design under test using the optimized set of parameters. In aspects, the method is repeated using the optimized set of parameters until a coverage percentage is reached, or in some implementations, full coverage may be reached. Some implementations of the method utilize black-box optimization through use of a Bayesian optimization algorithm.