You have your installation running and your first workload clusters are up and running? Then you can find more advanced guides for optimization and specific solutions in this section.
-
Ingress
From advanced configuration options for NGINX ingress controller, to running multiple ingress controllers, this sections provides some great depth around ingress.
-
Storage
Advanced usage of persistent volumes, especially for AWS users. About using the CSI driver with EBS, using Amazon's Elastic File System (EFS) and troubleshooting tips.
-
CoreDNS
Here we describe how you can customize the configuration of the managed CoreDNS service in your clusters
-
High-availability
How to increase the uptime of your workload clusters by using multiple availability zones for worker nodes, or using control plane nodes in several availability zones.
-
Spot instances/VMs
AWS spot instances and Azure spot virtual machines are a simple way to save on compute cost, if your use case permits it. Here you find documentation for using them with Giant Swarm Kubernetes clusters.
-
Node pools
A general description of node pools as a concept, it's benefits, and some details you should be aware of.
-
Automatic termination of unhealthy nodes
Unhealthy cluster nodes can lead to impaired workload reliability and wasted cluster resources. Here we explain how you can activate automatic termination of such nodes.
-
Cluster autoscaler
Here we describe how you can customize the configuration of the managed cluster autoscaler service in your workload clusters.
-
Cost-optimization
Cost is always a concern when running IT infrastructure. This document is a curated list of cost optimization tools that we have found to be valuable for our customers and ourselves.
-
Labelling clusters
Complete guide on using labels with workload clusters for the purpose of grouping, categorization and selection.
-
Authenticating with Azure AD
Tutorial on how to setup Authentication with Microsoft Azure Active Directory in kubectl.
-
Accelerated networking on Azure
A general description of the accelerated networking for Azure VMs feature and how it works on Giant Swarm clusters.
-
Egress IP address on Azure
How to reuse an existing public IP address for outgoing traffic of worker nodes on Azure.
-
Using GPUs
Here we explain how to make the GPU driver for Flatcar container linux available so that workloads can use GPUs.
-
Network pools
A general description of network pools as a concept, it's benefits, and some details you should be aware of.
-
Fine-tuning upgrade disruption
The level of disruption caused by cluster upgrades can be influenced per cluster. This article explains how to adjust the number of nodes that is updated simlutaneously, and the wait time between batches of nodes.
-
Custom admission controllers
Tutorial on how to create your own admission controller to enforce custom rules and policies in the resources created in your Kubernetes cluster.
-
Multi-account
By default, all your workload clusters run in the same cloud provider account. With multi-account support for AWS and Azure, you can define a specific cloud provider account to use per organization.
