We need more and more complex chips and SoCs for all new applications that use the latest technologies like AI. For example, Apple’s 5nm SoC A14 features 6-core CPU, 4 core-GPU and 16-core neural engine capable of 11 trillion operations per second, which incorporates 11.8 billion transistors, and AWS 7nm 64-bit Graviton2 custom processor contains 30 billion transistors. Designing such complex chips demands a standard and proven verification flow that involves extensive verification at every level, block to IP to Sub-system to SoC, using various verification methodologies and technologies.
In this article, let us walk you through various verification methodologies we use for verifying IPs, Sub-systems, and SoCs and explain why we need new methodologies/standards like PSS.