Whether you are just getting started with Datadog or you are long time users of the Datadog platform, there are always improvements and new features popping up to expand your observability practices. Through this blog we hope to shed some light on some of the tips and tricks we have learned through setting up Datadog observability in large Windows environments.

Choosing the right agent deployment method

• Manual installation of the Datadog agent can work very well during a testing phase or in small environments. In large production environments, it is necessary to have automation tooling to deploy and manage configuration at scale.
• Utilize existing tooling your team already uses to manage the environment. We have seen Ansible, PowerShell, SCCM, Bigfix and others work well for customers.
• Choose whichever deployment method works best for your team, but make sure you are able to easily update and change the configuration.

Think about what data you are collecting

• Remember the 3 pillars of observability (Metrics, Traces, and Logs).
• Datadog has a variety of different types of observability data you can collect. These have different benefits and price points depending on your license.
• Another benefit to keep in mind with the Datadog platform is that you can have everything in one tool. The more you consolidate into a single tool, the less you will have to context switch during outages.

Tips, Tricks and FAQ

• Tagging is critical to any Datadog environment. Standardize and iterate on your tagging strategy.
• By default, the Datadog agent installer will create a local user account ddagentuser to run the Datadog agent. The Datadog agent also supports domain accounts and Group managed service accounts (gMSA).
• How are you managing agent secrets? Consider using the Datadog Agent’s secret management feature to remove plaintext passwords and securely store your credentials.
• Datadog’s Fleet automation page is extremely helpful for managing agent versions and reviewing auto detected integrations that can be set up. Periodically review this to ensure you are up to date and collecting all the necessary data.
• Worried about overhead on the Datadog agent? Datadog has some example overhead calculations, as well as benchmarks for APM.

Ready to optimize your Datadog environment? Contact us today and unlock the true potential of your observability stack.