parca-dev/parca
Parca
Parca is an always-on continuous profiling platform that captures CPU and memory usage from running applications without any code modifications. It uses eBPF kernel-level tracing to automatically discover and sample stack traces across infrastructure, and provides a web-based flame graph dashboard for interactive performance analysis. Its label-based query engine lets users slice and aggregate profiling data across dimensions such as service, container, or region, using a Prometheus-style selector syntax.
Unlike basic profilers, Parca stores profile samples in a columnar format using Apache Arrow and Parquet, enabling fast aggregation and time-range filtering over large datasets. It supports differential flame graph comparison between two snapshots, with color-coded highlights showing where resource consumption increased or decreased. The platform integrates directly into Grafana dashboards for unified monitoring, and encodes the current analysis view—including flame graph parameters, label filters, and time range—into a stable, shareable URL.
Parca deploys profiling agents that automatically discover and profile containers on each node, sending data to a central server for storage and querying. It ingests profiles in the standard pprof format for interoperability with existing toolchains, and manages debug information through binary build ID extraction, local symbol extraction, and remote upload for symbol resolution. Multi-dimensional label-based queries and automated metadata storage complete the data management surface, making it possible to reconstruct function names and source locations from raw stack traces.
Features
- Continuous Profilers - Continuously samples CPU and memory usage across infrastructure with automatic target discovery.
- Debug Symbol Resolvers - Parca resolves raw stack traces to function names and source file lines using uploaded debugging information for accurate performance analysis.
- Profile Query Engines - Queries performance data by labels like service, container, or region to slice and aggregate profiles.
- Flame Graph Generators - Renders stack traces as flame graphs and icicle graphs in the browser, supporting click-to-drill-down and time-series overlay.
- Differential - Parca generates a color-coded differential flame graph that highlights code paths with increased (red) or decreased (green) resource consumption between two snapshots.
- Profile Query Engines - Selects and aggregates profiling data using multi-dimensional labels stored in a columnar database.
- Stores - Stores profile samples as columnar data using Apache Arrow and Parquet, enabling fast aggregation and time-range filtering over large datasets.
- Zero-Instrumentation Profilers - Parca continuously collects CPU and memory profiles using kernel-level tracing and eBPF, with automatic target discovery and no code instrumentation required.
- Profiling Server Deployments - Parca provisions a server component that receives, stores, and serves profiling data via a web interface and API for querying.
- Data Collection Agents - Parca deploys agents that automatically discover and profile containers on each node, sending collected data to a central server for aggregation and analysis.
- eBPF Profilers - Captures stack traces from user and kernel space using eBPF without modifying application code.
- eBPF Profilers - Continuously collects CPU and memory profiles from production systems using eBPF kernel tracing.
- Flame Graph Visualizers - Renders interactive flame graphs and differential comparisons to highlight performance changes.
- pprof Profile Converters - Ingests and exports profiles in the standard pprof format for interoperability with existing toolchains.
- Profile Format Ingestion - Parca ingests profiles in the standard pprof format, allowing interoperability with existing profiling tools and ecosystems across different programming languages.
- Debug Symbol Extractors - Resolves raw memory addresses to human-readable function names and source line numbers by extracting and caching debug symbols from binaries.
- Profiling Metadata Stores - Parca stores function names, file line numbers, and mapping details in a deduplicated key-value store for efficient symbol resolution during analysis.
- Profiling Data Dashboards - Parca embeds continuous profiling data into Grafana dashboards for integrated monitoring and interactive exploration alongside other metrics.
- Profiling Target Filters - Parca configures the profiling agent to only collect data from targets that match specified Kubernetes label selectors, reducing noise.
- Profiling View State Encoders - Encodes current flame graph view parameters, label filters, and time range into a stable URL for sharing exact profiling views.
- Observability and Monitoring - Continuous profiling for CPU and memory usage.
Star history
How this analysis was created: This summary and feature list were written by an AI model that read the project's README and public documentation pages. Each feature links to the documentation it came from; stars, license and language come straight from the GitHub API. The model does not read the source code, and the analysis is refreshed when the project is re-analysed. Learn more on our About page.
