Datadog

Bidirectional integration between Datadog and Knowledge Tree for enriched observability and topology-aware monitoring.

Overview

The Datadog integration connects your monitoring telemetry with the knowledge graph. Resources discovered by Knowledge Tree are enriched with Datadog metrics, monitors, and alert data. Conversely, Datadog dashboards and monitors can reference graph data to provide topology-aware observability.

Metric correlation

Knowledge Tree automatically correlates Datadog metrics with discovered resources. When viewing a resource, you can see its key performance indicators pulled from Datadog:

CPU and memory utilization -- for compute resources (EC2, VM, pods)
Request latency and error rates -- for load balancers, API gateways
Connection pool and query latency -- for databases (RDS, Cloud SQL)
Disk and network I/O -- for storage resources

Resource tagging

Knowledge Tree tags are synchronized to Datadog, and vice versa. This ensures consistent resource grouping across both platforms:

Tag sync direction -- configurable (bidirectional, KT to Datadog, or Datadog to KT)
Conflict resolution -- last-write-wins or manual preference
Tag-based scoping -- Datadog tags become Knowledge Tree scope filters

Dashboard links

When viewing a resource in the Knowledge Tree UI, contextual links to relevant Datadog dashboards are automatically generated. Similarly, Datadog dashboard widgets can link back to the Knowledge Tree resource detail page.

Monitor enrichment

Datadog monitor notifications can be enriched with Knowledge Tree data via webhooks. When a monitor triggers an alert, the payload includes:

Resource graph -- link to the resource in the knowledge graph
Dependency information -- what depends on the monitor's resource
Recent changes -- what changed on the resource in the last 24 hours
Runbook links -- relevant operational runbooks

Incident response workflow

Combined with the PagerDuty integration, Datadog monitor alerts flow through Knowledge Tree for enrichment before reaching your on-call team, reducing MTTR by providing context directly in the alert.

Configuration

# config.yaml
integrations:
  datadog:
    api_key: ${DATADOG_API_KEY}
    app_key: ${DATADOG_APP_KEY}
    site: datadoghq.com
    metric_correlation:
      enabled: true
      metrics:
        - "aws.ec2.cpuutilization"
        - "aws.rds.database_connections"
    tag_sync:
      enabled: true
      direction: bidirectional
    enrichment:
      monitors: true
      dashboards: true