Node Pools

With the release of GSK Version 1.30.3-gs0, support for Node Pools has been introduced, allowing users to create dedicated groups of nodes optimized for different types of workloads. This feature provides enhanced flexibility and control over resource allocation, enabling more efficient workload management in distributed environments.

Overview

Node Pools in GSK Version 1.30.3-gs0 allow for the creation and management of multiple groups of nodes, each configurable to suit specific workload requirements. This enables fine-grained control over infrastructure resources, allowing you to:

  1. Scale worker nodes within a pool to meet workload demands.
  2. Configure resources such as CPU, memory (RAM), and storage for each node pool, including advanced storage options like Rocket Storage.
  3. Add or remove pools dynamically to optimize the cluster’s performance based on the workload.

By segmenting workloads across different node pools, you can ensure that resource-intensive tasks are isolated from lightweight processes, improving performance, resource utilization, and operational efficiency.

Limitation

While Node Pools in GSK provide a robust solution for managing workloads, there are some current limitations to be aware of:

  • No Support for Taints and Tolerations: At present, GSK Version 1.30.3-gs0 does not support the use of taints and toleration. This means node pools cannot yet enforce exclusive workload scheduling policies. Without this feature, workloads may not be explicitly restricted to specific node pools, limiting more granular control over workload isolation.

However, users can manage workload placement by leveraging predefined labels (e.g., the provider label with the node pool name, which can be viewed using kubectl describe node) or by manually applying custom labels to specific nodes. Additionally, taints can be used to control pod scheduling, ensuring workloads are assigned to the appropriate nodes.

It is recommended to use predefined labels for greater reliability and persistence, as these labels remain intact even after a node is restarted or recycled.

Furthermore, upcoming releases will include enhanced support for taints and tolerations, providing additional flexibility in workload placement and scheduling.