Seeking the Yield Barrier: High-Dimensional SRAM Evaluation Through Optimal Manifold

TL;DR

This paper introduces OPTIMIS, a novel high-dimensional SRAM yield estimation method combining optimal manifold theory, neural flows, and importance sampling, achieving significant improvements in efficiency and accuracy.

Contribution

It develops the optimal manifold concept for yield estimation, integrating neural coupling flows with importance sampling to enhance high-dimensional SRAM analysis.

Findings

Up to 3.5x efficiency improvement over state-of-the-art methods.

Up to 3x accuracy improvement in yield estimation.

Robust and consistent performance across high-dimensional SRAM evaluations.

Abstract

Being able to efficiently obtain an accurate estimate of the failure probability of SRAM components has become a central issue as model circuits shrink their scale to submicrometer with advanced technology nodes. In this work, we revisit the classic norm minimization method. We then generalize it with infinite components and derive the novel optimal manifold concept, which bridges the surrogate-based and importance sampling (IS) yield estimation methods. We then derive a sub-optimal manifold, optimal hypersphere, which leads to an efficient sampling method being aware of the failure boundary called onion sampling. Finally, we use a neural coupling flow (which learns from samples like a surrogate model) as the IS proposal distribution. These combinations give rise to a novel yield estimation method, named Optimal Manifold Important Sampling (OPTIMIS), which keeps the advantages of the surrogate and IS methods to deliver state-of-the-art performance with robustness and consistency, with up to 3.5x in efficiency and 3x in accuracy over the best of SOTA methods in High-dimensional SRAM evaluation.