Datadog

Datadog is our primary observability tool. Logs, metrics, and traces from our products are forwarded to Datadog for deeper analysis, monitoring, and visualization.

The main reasons we chose it over other alternatives are:

• It is a mature product with extensive documentation and good technical support for customers.
• It centralizes many of the tools we use into a single solution (bug tracking system, alerts, custom dashboards, etc).
• It is a constantly evolving product with new features being added regularly like Cloud SIEM or Kubernetes monitoring.
• Many ways to visualize information, making it easier to find issues and identify patterns.
• Many integrations available with cloud providers and third-party applications.

Alternatives

We have considered other observability tools like:

• Bugsnag. Main used service was the bug tracking system. We migrated it to Datadog Error Tracking looking for centralization.
• LogRocket. Main used service was the real user monitoring. We migrated it to Datadog Real User Monitoring looking for centralization.
• Atatus. Simpler and cost-effective alternative. During trial, we migrated because Atatus didn’t support Logs observability and K8s infra monitoring. Also, it lacked of compliance features like a Cloud SIEM or bug tracking system.
• Coralogix. Simpler and cost-effective alternative. During trial, Coralogix was complex to use by devs, it didn’t have enough integrations and dashboards are not very helpful. Also, it lacked of compliance features like a Cloud SIEM or bug tracking system.
• New Relic. Similar to Datadog. Cost model is complex and includes a per-user pricing. Also, they had very bad support for asyncio when we were migrating to Starlette.

Usage

We use Datadog for:

• Indexed logs, custom metrics and platform tracing.
• Custom dashboards and monitoring rules.
• SLOs and alerts.
• Bug tracking system.
• Real User Monitoring.
• Infrastructure monitoring.
• Cloud SIEM for threat detection from third-party sources.
• Compliance.