A hardware accelerator chip designed specifically by Google for machine learning training and inference workloads, optimized for massive matrix operations, as an alternative or complement to GPUs.
In GENAI-L (Domain 2) it appears within Google Cloud's "AI-optimized infrastructure" alongside GPUs and AI Hypercomputer. A key differentiator for Google versus other providers: its own silicon design (TPU) vertically integrated with its models.
TPU (Tensor Processing Unit) is a concept that can show up on the exam. Check the full guide to see it in context.