DataDog/datadog-agent
Datadog Agent
The Datadog Agent is an infrastructure monitoring agent and host telemetry collector. It functions as a background process that gathers system metrics and application health data to send to a centralized monitoring platform.
The project operates as a plugin-based metric collector, using a modular system of independent check scripts to gather data from various third-party services and applications. It serves as a remote telemetry transmitter, providing a pipeline to stream infrastructure and system information to a remote analysis and alerting backend.
Its capabilities cover application performance monitoring, host resource tracking, and infrastructure performance monitoring. The agent collects low-level system telemetry from the operating system kernel and filesystem while aggregating application-level performance data to identify service degradation.
Features
- System Telemetry Collectors - Gathers performance metrics from kernel interfaces and system files for monitoring purposes.
- Kernel-Level Operations - Interacts directly with the operating system kernel and filesystem to extract low-level hardware and resource metrics.
- Metric Collection Plugins - Uses a modular system of independent check scripts to gather specific data from third-party services and applications.
- Application Performance Monitoring - Tracks runtime errors and performance bottlenecks in distributed software to ensure stability.
- Modular Check Collectors - Implements a modular system using independent check scripts to gather data from various third-party services.
- System Metrics Collection - Gathers native system-level performance metrics from the operating system kernel and filesystem.
- Monitoring Agents - Acts as a lightweight software agent on remote hosts to collect and push system health metrics to a central server.
- System Resource Tracking - Monitors server-level hardware and network resource utilization including CPU, memory, and disk.
- Application Diagnostic Data - Collects stability and performance diagnostic telemetry from running applications to identify bottlenecks.
- Application Health Monitors - Tracks the operational health and performance metrics of running software applications.
- Telemetry Collection and Aggregation - Provides the transport and streaming infrastructure to send telemetry data to a remote platform for analysis.
- Infrastructure Health Monitoring - Provides real-time observation of server hardware and system resource stability to ensure availability.
- Remote Telemetry Dashboards - Provides the telemetry stream that populates remote monitoring dashboards for system health and performance visualization.
- System Health Monitors - Tracks real-time system metrics, hardware status, and resource usage of the host operating system.
- Telemetry Transmission Pipelines - Implements a secure pipeline for streaming collected infrastructure telemetry to a remote analysis and alerting backend.
- Auto-Discovery Mechanisms - Automatically identifies running services and monitors by scanning system processes and reading local configuration files.
- Process Isolation Architectures - Runs separate processes for the core agent and specialized collectors to isolate failures and manage resources.
- Metric Aggregators - Collects individual metrics over the network and summarizes them for centralized backend delivery.
- Periodic Metric Polling - Triggers metric collection tasks at defined intervals using a central scheduler to ensure consistent time-series data.
