The Role of Cloud Computing in a Modern Reverse Engineering Workflow at the 5nm Node and Beyond

Abstract

A modern reverse engineering (RE) workflow contains many challenges, especially as process nodes drop below the 5nm node. With increased complexity, more circuitry is packed into a smaller area, requiring large quantities of raw data collected and subsequently processed to help reconstruct the original schematics. By leveraging inexpensive cloud computing, orders of magnitude improvement in throughput were achieved for 2D image registration and high quality image segmentation was achieved using machine learning.