This module offers an overview of the course and outlines the learning objectives.

Course introduction

Course overview

This module details the specialized networking requirements for AI workloads compared to traditional web applications. It covers the specific bandwidth and latency demands of each pipeline stage—from ingestion to inference—and analyzes the "rail-aligned" network architectures of Google Cloud's A3 and A4 GPU machine types designed to maximize "Goodput."

Role of networking in AI pipeline

AI Hypercomputer networking for ML workloads

Networking for GPU machine types

Quiz Question 1

Quiz Question 2

Introduction

This module details strategies for efficiently moving massive datasets into the cloud. It covers the use of the Cross-Cloud Network and Cloud Interconnect to establish high-bandwidth pipelines, and outlines configuration best practices—such as enabling Jumbo Frames (MTU)—to reduce protocol overhead and optimize throughput.

Overview

Data ingestion over the Cross-Cloud Network

Networking for data ingestion

This module details the critical role of low-latency networking in distributed model training. It covers the necessity of Remote Direct Memory Access (RDMA) for gradient synchronization, the benefits of Google's Titanium offload architecture in freeing up CPU resources, and the topology choices required to scale clusters without bottlenecks.

Networking considerations for training

Networking for distributed training

Titanium's modern offloads architecture

Networking for AI training

This module details the networking challenges specific to Generative AI inference, such as bursty traffic and long-lived connections. It covers optimizing Time-to-First-Token using the GKE Inference Gateway and "Queue Depth" routing, while also addressing best practices for network reliability and Identity and Access Management (IAM).

Networking for generative AI inferencing

Optimizing network reliability for AI inference

Optimizing inference with GKE Inference Gateway

AI networking best practices

Congratulations and next steps

Quiz Question 3

Networking for inference

Course Resources

Welcome to the "AI Infrastructure: Networking Techniques" course. In this course, you'll learn to leverage Google Cloud's high-bandwidth, low-latency infrastructure to optimize data transfer and communication between all the components of your AI system. By the end, you'll grasp the critical role networking plays across the entire AI pipeline from data ingestion and training to inference and be able to apply best practices to ensure your workloads run at maximum speed.

AI Infrastructure Networking Techniques

Design and deploy high-performance AI/ML solutions using Google Cloud's AI Hypercomputer, GPUs, TPUs, Compute, and Google Kubernetes Engine.

Optimizing network reliability for AI inference

Interaktive praktische Übung