Manuel Cataldo
AI-Optimized Design Verification: Comparative Analysis.
Rel. Guido Masera. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Elettronica (Electronic Engineering), 2026
Abstract
Digital verification teams increasingly rely on large constrained–random regressions to achieve coverage closure and find bugs early. While robust orchestration platforms ensure repeatability and visibility, static test selection often leads to redundancy and slow progress on rare or unhit coverage targets. This thesis presents a comparative analysis of using Synopsys VC Execution Manager (ExecMan) alone versus using ExecMan together with the VSO.ai Regression Optimizer plugin. Grounded in official user guides and common industrial practice, we define a simple framework to compare workflows, coverage trajectories, stimulus diversity, and resource efficiency. ExecMan alone provides strong execution management, coverage merging, and reporting, but leaves test selection to manual curation or broad, static lists.
By contrast, ExecMan + VSO.ai adds a learning–based optimizer that builds a compact Hit Matrix from prior runs and recommends the “next best” tests dynamically, focusing on high–value stimuli and rare coverage bins
Relatori
Anno Accademico
Tipo di pubblicazione
Numero di pagine
Informazioni aggiuntive
Corso di laurea
Classe di laurea
Aziende collaboratrici
URI
![]() |
Modifica (riservato agli operatori) |
