apache/skywalking
Skywalking
SkyWalking is an application performance monitoring system and observability platform designed to collect and analyze metrics, traces, and logs from distributed microservices. It functions as a distributed tracing platform and a telemetry data pipeline that ingests and aggregates observability data from various language agents.
The project features an AI-powered anomaly detector that uses machine learning to calculate metric baselines and identify irregular URI patterns. It includes an eBPF performance profiler for diagnosing CPU and network bottlenecks at the kernel level and generates interactive service topology visualizations to map dependencies between distributed services.
The system covers broad capability areas including agent-based data collection, log data processing, and performance alerting. It employs a multi-backend storage abstraction and a service provider interface to support custom data receivers and storage backends.
The project provides tooling for backend infrastructure orchestration using container composition and a command line interface for system administration.
Features
- Application Performance Monitoring - Collects and analyzes metrics, traces, and logs to ensure application health and stability in production.
- Application Performance Monitoring Platforms - Provides a centralized platform for aggregating runtime metrics and error data across distributed software environments.
- Telemetry Data Pipelines - Implements a high-performance engine for the ingestion, processing, and aggregation of multi-source observability data.
- Topology Visualizers - Generates interactive maps representing relationships and dependencies between distributed microservices.
- Distributed Tracing - Tracks requests across distributed systems using unique identifiers to analyze performance and visualize execution flow.
- Distributed Request Tracers - Maps request flows and service dependencies across complex microservice architectures to identify bottlenecks.
- Telemetry Collection and Aggregation - Provides infrastructure for the transport, streaming, and unification of telemetry data from diverse agents.
- Performance Profiling Tools - Diagnoses CPU and network performance bottlenecks using eBPF-powered system profiling.
- eBPF Profilers - Captures low-level kernel and network events via eBPF to diagnose system bottlenecks.
- Service Dependency Mapping - Builds interactive maps of microservice dependencies by analyzing distributed trace identifiers.
- Telemetry Agents - Employs language-specific agents to collect performance data and forward it to a central collector.
- Telemetry Ingestion - Ingests and aggregates metrics, traces, and logs from diverse ecosystems using a unified pipeline.
- eBPF Profilers - Diagnoses CPU and network performance bottlenecks at the kernel level using extended Berkeley Packet Filter tools.
- Data Collection Agents - Gathers metrics, logs, and traces from distributed systems using language-specific runtime agents.
- Telemetry Query Languages - Executes specialized queries against processed metrics to retrieve performance and reliability data.
- Provider Integrations - Allows integration of new data receivers and storage backends via modular provider interfaces.
- Telemetry Processing Languages - Processes logs and metrics through a programmable pipeline using a domain-specific language.
- Extensible Processing Pipelines - Enables custom processing logic for observability and log data through extensible analysis pipelines.
- Modular Provider Interfaces - Provides a modular provider interface for integrating custom data receivers and storage backends at runtime.
- Storage Abstraction Layers - Implements a storage abstraction layer to decouple the telemetry engine from specific database backends.
- Alerting Systems - Configures alarm rules based on services or APIs to route notifications to third-party monitoring systems.
- Anomaly Detection - Uses machine learning to calculate metric baselines and automatically identify irregular system patterns.
- API Performance Monitoring - Provides query interfaces to fetch and analyze API-level performance metrics and health data.
- Log Processing Pipelines - Analyzes and structures unstructured log data using a dedicated processing language to extract metrics.
- Telemetry Protocols - Uses standardized protocols to transmit performance and reliability metrics to monitoring backends.
- Log Aggregation Pipelines - Collects, formats, and indexes distributed system logs through a high-performance processing pipeline.
- Observability Backends - Application performance monitoring system for distributed systems.
- Distributed Tracing - Observability platform for microservices and cloud-native environments.
- Tracing And Profiling - APM system for microservices and cloud-native apps.
