Monitor Your T2A-Powered GKE Workloads With Datadog | Datadog

Monitor your T2A-powered GKE workloads with Datadog

Author Rachel Groberman
Author Mallory Mooney

Published: 7月 13, 2022

Arm processors have become increasingly popular in recent years, providing energy-efficient, cost-effective processing power to both mobile and cloud computing ecosystems. As a part of this growth, more and more organizations are choosing to leverage the many benefits of Arm-based architectures for their containerized workloads. Today, Google Cloud announced its Arm-based Tau T2A virtual machines (VMs), which you can also use to run workloads in Google Kubernetes Engine (GKE).

Datadog already provides complete visibility into your GKE environment, including any nodes running on Arm-based VMs. This visibility is crucial for migrating workloads over to Tau T2A machines, as it enables you to easily compare costs and performance between Arm- and x86-based architectures.

Automatically visualize data across your GKE clusters

You can start monitoring your Arm-powered workloads in Datadog by enabling the GKE integration and deploying the Datadog Agent onto your clusters. Datadog’s built-in integration dashboard allows you monitor key data across nodes, such as CPU utilization, so you can determine which ones could benefit from leveraging the new T2A machines.

View the status of all Arm-based nodes with Datadog's GKE dashboard.
Monitor all of your GKE nodes, including those powered by Arm processers, with Datadog's GKE dashboard.

Analyze performance for Arm nodes

Arm-based VMs are designed to optimize performance and cost, so Datadog can help you ensure that newly provisioned nodes are appropriately configured to handle traffic as expected. With Datadog APM, you can collect traces from nodes running on different architectures in order to compare performance data such as request latency, error rate, and throughput.

Monitor performance between x86 and Arm nodes with Datadog APM.
Analyze trace data for services running on Arm nodes alongside their x86-based counterparts.

For example, your GKE workloads will experience typical peaks and drops in request throughput based on application usage, but any sustained spikes could indicate that you have not allocated enough resources to nodes. You can leverage Datadog’s Live Container view to review configurations for specific Arm nodes to determine if they are properly rightsized for processing incoming requests. By pivoting between traces and node-specific data, you can easily monitor performance improvements in newly deployed Arm nodes before, during, and after a migration.

Start monitoring your Arm-powered GKE workloads

Datadog provides complete visibility into your GKE workloads, including those powered by Google Cloud’s Tau T2A VMs. You can easily visualize performance data across your Kubernetes environment and verify that your containerized applications are benefiting from the Arm-based architecture’s capabilities. If you’re a Datadog customer, you can enable the GKE integration and deploy the Agent to your nodes today. If you’re not yet a Datadog customer, you can get started with a .