AI Infrastructure: Orchestration and Automation

AI Infrastructure: Orchestration and Automation

Instructor: Google Cloud Training

Access provided by Universidad Autonoma Metropolitana

5 modules

Gain insight into a topic and learn the fundamentals.

Intermediate level

Some related experience required

4 hours to complete

Flexible schedule

Learn at your own pace

5 modules

Gain insight into a topic and learn the fundamentals.

Intermediate level

Some related experience required

4 hours to complete

Flexible schedule

Learn at your own pace

What you'll learn

Recognize how GKE uses containerization to eliminate environment inconsistencies and optimize GPU/TPU resources across the AI lifecycle.
Identify core Kubernetes AI constructs and explain how Terraform, Cluster Toolkit, and Cluster Director automate cluster setup.
Explain how Kueue uses custom resources (ResourceFlavor, ClusterQueue, LocalQueue) to prevent job starvation and manage traffic.
Differentiate between DWS Flex Start and Calendar modes, and describe how Topology-Aware Scheduling (TAS) reduces latency.

Skills you'll gain

Tools you'll learn

Details to know

Shareable certificate

Add to your LinkedIn profile

Assessments

1 assignment

Taught in English

See how employees at top companies are mastering in-demand skills

Learn more about Coursera for Business

logos of Petrobras, TATA, Danone, Capgemini, P&G and L'Oreal

There are 5 modules in this course

This course covers the automation and operational management of GKE clusters optimized for distributed AI and machine learning workloads. You will learn to provision accelerator resources, configure advanced workload queueing, mitigate capacity constraints, and orchestrate large-scale distributed jobs.

This module defines the purpose of the course.

What's included

1 plugin

Examine AI infrastructure challenges and explore core Kubernetes components to learn how GKE optimizes the full AI lifecycle. Additionally, explore various tools available to automate cluster creation.

What's included

4 plugins

Optimize multi-tenant AI workloads on GKE using Kueue for job queuing, Dynamic Workload Scheduler (DWS) for GPU/TPU provisioning, Topology-Aware Scheduling (TAS) for low latency, and custom Compute Classes (CCC) for automated compute fallbacks.

What's included

6 plugins

The module dives into the KubeRay operator architecture and its custom resources (RayCluster, RayJob, and RayService). We will also explore how to orchestrate distributed Ray workloads on GKE.