This document provides a consolidated overview of best practices for designing, building, and operating applications on Google Kubernetes Engine (GKE). Following these recommendations helps you optimize for cost, performance, security, and reliability. Each entry links to more detailed documentation for specific topics.
Select your GKE category:
Or, search for a best practice:
| Category | Best practices | Summary |
|---|---|---|
|
AI and ML workloads
|
Optimize AI/ML workload efficiency | Maximize resource efficiency for AI/ML tasks on GKE. |
|
AI and ML workloads
|
Inference workloads | Run machine learning inference workloads on GKE. |
|
AI and ML workloads
|
Autoscale LLM inference with GPUs | Automatically scale large language model (LLM) inference using GPUs on GKE. |
|
AI and ML workloads
|
Autoscale LLM inference with TPUs | Automatically scale large language model (LLM) inference using TPUs on GKE. |
|
AI and ML workloads
|
Optimize LLM inference with GPUs | Optimize LLM inference with GPUs on GKE. |
|
AI and ML workloads
|
Batch processing platform | Build and operate a batch processing platform on GKE. |
|
Cost optimization
|
Run cost-effective Kubernetes applications | Reduce the operational costs of Kubernetes applications on GKE. |
|
Databases
|
Database options | Choose and manage database solutions for GKE applications. |
|
Networking
|
Networking | Configure and manage network connectivity for GKE. |
|
Operations
|
Upgrading clusters | Learn how to perform smooth and reliable GKE cluster upgrades. |
|
Operations
|
CI/CD for GKE | Implement continuous integration and delivery pipelines for GKE applications. |
|
Reliability and scalability
|
Scalability | Learn principles and techniques for scaling applications on GKE. |
|
Reliability and scalability
|
Plan for scalability | Learn strategies for designing scalable GKE environments. |
|
Reliability and scalability
|
Plan large GKE clusters | Learn how to architect and manage large-scale GKE clusters. |
|
Reliability and scalability
|
Best practices for running HPC workloads | Best practices to optimize running high performance computing (HPC) workloads on GKE. |
|
Reliability and scalability
|
Plan large workloads | Deploy and manage resource-intensive applications. |
|
Security
|
Harden your GKE cluster | Enhance the GKE security posture of your GKE clusters. |
|
Security
|
Plan RBAC policies | Define role-based access control to manage permissions. |
|
Security
|
Enterprise multi-tenancy | Run multiple tenants securely on a single GKE cluster. |
