Process variability at advanced technology nodes has become a key challenge for IC designers. A new generation of tools are required that provide efficient and reliable solutions for analog, RF, standard cells, IO and memory designers beyond traditional Monte Carlo analysis. A comprehensive suite of new tools that allow the designer to accurately address statistical process variations and to make the right design decision upfront is needed. The VarMan tool suite ensures that designers do not need to be expert statisticians to understand and optimize the impact of process variations on their design.
Relying on advanced machine learning approaches, VarMan’s breakthrough technology delivers up to 30X simulation performance compared to traditional Monte Carlo simulation, while preserving accuracy and precision. For designs with high-? requirements, VarMan makes precision-driven analysis easily achievable for design teams. Designed to simplify the setup of characterization projects, VarMan requires little effort from designers to setup and run. Our full-chip XMA memory yield analysis uses VarMan’s innovative fast Monte Carlo and robust high-sigma kernels for a fast fail detection and yield estimation of an entire memory.
Machine Learning in the EDA-Specific Domain – 20 Years in the Making
Memory Statistical Characterization Solution with VarMan
Standard Cell Statistical Characterization with VarMan
Statistical Analysis Flow for Analog Design with VarMan
Variation aware design for advanced nodes and low power technologies