Open-source alternatives to Deepflow

Pyroscope is a continuous profiling platform designed to collect, store, and visualize application performance data. It functions as an application performance management suite that tracks historical resource usage to identify bottlenecks and detect performance regressions over time. The platform distinguishes itself through its use of kernel-level instrumentation and dynamic runtime hooks, which allow for performance monitoring without requiring manual code modifications or application restarts. It employs a sidecar agent architecture to offload telemetry processing, utilizing delta-encoded

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 inter

Pinpoint is a distributed application performance monitoring and tracing system. It functions as an application performance monitor and topology visualizer designed to analyze the execution behavior of large-scale distributed applications. The system uses bytecode instrumentation to monitor applications without requiring changes to the original source code. It captures call stacks and request flows across interconnected services to visualize system dependencies and generate real-time architectural maps of communication patterns. The platform covers a broad range of observability capabilities

Hubble is an eBPF-based Kubernetes observability platform designed for network monitoring, security auditing, and flow inspection. It provides deep visibility into containerized traffic and cluster security by utilizing kernel-level hooks to collect network events. The system features a service map for visualizing communication patterns and dependencies between microservices and external endpoints. It incorporates identity-based flow labeling to track network traffic using Kubernetes labels rather than volatile IP addresses. The platform covers a broad range of monitoring capabilities, inclu

Inspektor Gadget is an eBPF observability toolset and program framework designed for tracing Linux systems and debugging Kubernetes nodes. It provides a suite of tools to collect kernel-level telemetry and export system metrics via the OpenTelemetry standard. The project distinguishes itself by packaging inspection tools as OCI-compliant container images, allowing for standardized distribution and deployment across clusters and hosts. It employs a modular data processing pipeline that utilizes WebAssembly modules to transform and filter telemetry, and leverages Compile Once Run Everywhere for

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 A

GreptimeDB is a distributed, open-source time-series database built for unified observability. It stores and queries metrics, logs, and traces together in a single columnar engine, supporting both SQL and PromQL for analysis. The database is designed as a Kubernetes-native operator with a decoupled compute and storage architecture, enabling horizontal scaling and multi-region deployment. What distinguishes GreptimeDB is its role as a multi-protocol ingestion gateway, accepting data through OpenTelemetry, Prometheus Remote Write, InfluxDB, Loki, Elasticsearch, Kafka, and MQTT protocols without

This project is a comprehensive technical guide and diagnostic manual for analyzing memory, performance, and asynchronous behavior within Node.js applications. It provides detailed methods for asynchronous tracing, memory diagnostics, and performance analysis to resolve runtime errors and execution bottlenecks. The resource distinguishes itself by covering advanced diagnostic workflows, including the use of flame graphs for CPU profiling, the capture and comparison of heap snapshots for memory leak detection, and the mapping of asynchronous call stacks. It also provides technical guidance on

SkyWalking is a comprehensive observability stack and application performance monitoring platform. It functions as a distributed tracing system and an AI application monitor, providing a centralized suite for collecting and analyzing logs, metrics, and traces to maintain the health of containerized architectures. The platform distinguishes itself through a service topology visualizer that renders interactive maps of infrastructure dependencies and communication patterns. It also includes specialized capabilities for generative AI workflow observation to track the execution flow and performanc

This repository contains the specifications for a standardized observability data standard and telemetry instrumentation framework. It defines unified protocols for how telemetry data is instrumented, propagated, and formatted to ensure interoperability across distributed systems. The specification provides a vendor-neutral approach to observability, allowing telemetry data to be exported to various backends without locking into a single provider or requiring changes to application source code. It establishes a consistent protocol for capturing request flows across service boundaries to corre

BCC is an eBPF development toolkit and tracing framework used for monitoring and analyzing the Linux kernel. It functions as a performance analysis tool and debugging utility to capture system events, measure kernel latency, and provide network observability. The project distinguishes itself by providing a build system that integrates with LLVM to compile C-like code into BPF bytecode at runtime. It utilizes BPF Type Format data for relocations to maintain cross-kernel compatibility and extracts kernel headers to ensure the generated programs match the specific kernel version. The toolkit co

OpenObserve is a unified observability data platform designed to ingest, store, and analyze logs, metrics, and traces. It functions as a cloud-native monitoring tool that centralizes telemetry from diverse sources, including standard collectors and cloud service providers, into a single, scalable system. By utilizing a columnar storage engine backed by object storage, the platform enables efficient long-term data retention and high-performance analytical querying. The platform distinguishes itself through deep integration with artificial intelligence, allowing users to query data using natura

This project provides an observability data pipeline designed to collect, transform, and route logs, metrics, and traces from diverse sources into standardized formats for analysis. It operates as a plugin-based component architecture using modular receivers, processors, and exporters to move telemetry data through sequential processing chains. The system utilizes an interface-driven component model that allows for interchangeable connectors and community-contributed extensions. It distinguishes itself through a domain-specific language for telemetry filtering, metadata-based resource attribu

VictoriaMetrics is a high-performance, scalable time series database and observability platform designed for long-term storage and analysis of metric, log, and trace data. It functions as a unified backend for monitoring ecosystems, offering full compatibility with industry-standard protocols and query languages. The system is built to handle massive data volumes through a distributed architecture that supports horizontal scaling and efficient data lifecycle management. The platform distinguishes itself through a storage engine that utilizes consistent hashing for data sharding and log-struct

Quarkus is a Kubernetes-native Java framework designed for building high-performance, memory-efficient applications. It utilizes ahead-of-time native compilation to transform Java code into standalone, optimized binaries that eliminate the need for a virtual machine, enabling rapid startup and reduced memory consumption. By performing code augmentation during the build phase, it shifts heavy processing tasks away from runtime, ensuring that applications are optimized for cloud-native environments. The framework distinguishes itself through a unified approach to reactive and imperative program

This project is a deep learning framework designed for constructing, training, and deploying neural networks across diverse hardware environments. It functions as a high-performance tensor computation library that provides both imperative and symbolic programming interfaces, allowing developers to balance flexible, step-by-step model building with the efficiency of compiled computation graphs. The framework distinguishes itself through a hybrid execution engine that integrates declarative graph compilation with imperative runtime logic. It supports scalable, distributed training across multip

OpenTelemetry Go is a framework for generating and collecting distributed traces, metrics, and logs from Go applications. It provides a standardized telemetry instrumentation API for adding observability markers to code and a corresponding SDK for processing and emitting these signals. The project utilizes a configurable observability pipeline to sample and export telemetry data to external backends using the OTLP wire protocol. It features a pluggable export system and a separation between the public API and the SDK implementation, allowing telemetry to be routed to third-party platforms wit

Grafana Tempo is a high-scale distributed tracing backend and columnar trace database. It serves as an observability data store that persists and queries spans and traces using OpenTelemetry standards, allowing for the analysis of request flows across microservices. The system distinguishes itself by using an object-store based backend with columnar Parquet storage. This architecture enables efficient attribute searching and large-scale data retrieval through dedicated attribute columnization and block-based data partitioning. It includes a specialized TraceQL query engine for filtering trace

This project is a performance analysis tool for Ruby applications using the Rack interface. It monitors request execution times and resource usage, serving as a profiler for web applications to measure latency and identify bottlenecks. The tool provides specific analyzers for database query performance, memory allocations, and garbage collection statistics. It generates call stack flamegraphs to visualize time distribution across methods and renders speed badges and timing metrics directly onto HTML pages. The system covers broader performance profiling capabilities including custom code blo

Samply is a cross-platform CPU sampling profiler and performance analysis utility. It consists of a command-line tool for recording process stack traces at regular intervals and a visual interface for analyzing the resulting execution data. The system includes a debug symbol resolver that maps raw memory addresses to human-readable function names using local or remote symbol information. It transforms recorded execution data into flame graphs and timelines to pinpoint function-level hotspots. The tool provides capabilities for CPU execution recording, stack unwinding, and symbol resolution a

Async-profiler is a sampling profiler for Java applications that tracks CPU time and stack traces across execution frames to identify performance bottlenecks. It is designed to capture profiling data without introducing timing bias. The project provides capabilities for JVM memory analysis to locate native and heap allocation hotspots and memory leaks. It also includes system contention analysis to identify resource bottlenecks through the tracking of contended locks and hardware performance counters. The tool converts raw profiling data into visual performance reports, including interactive

Falco is an eBPF runtime security monitor and cloud native detection engine that identifies abnormal behavior and security threats across hosts and containers. It functions as a Linux kernel event auditor, capturing system calls and kernel events in real-time to detect malicious activity. The system distinguishes itself through a rule-based threat detection model that evaluates system activity against a library of community-maintained rules and custom security definitions. It enriches raw kernel events with container and Kubernetes metadata to provide observability into isolated environments

Cat is a distributed application performance monitoring tool and tracing framework designed to track transactions, latency, and health across distributed services. It functions as a Kubernetes-native monitoring stack that utilizes multi-language monitoring clients and a real-time alerting system to maintain system visibility. The system provides monitoring clients for Java, Go, Python, Node.js, and C++ to collect performance metrics and trace data. It distinguishes itself by sampling request flows to record call chains and identify bottlenecks, while using a monitoring engine to trigger immed

React Scan is a diagnostic utility and performance auditor designed to monitor the rendering lifecycle of components within user interfaces. It functions as an automated analysis tool that tracks component re-render cycles and execution timing to identify performance bottlenecks in real time. The tool distinguishes itself by providing visual feedback through a persistent overlay injected directly into the application. By instrumenting the reconciliation process and observing component state and props, it highlights specific rendering patterns that contribute to performance degradation. This

Tracy is a real-time performance profiling framework for C and C++ applications. It provides a software instrumentation library that captures high-resolution telemetry data, which is then visualized through a separate graphical interface to identify bottlenecks and resource allocation issues. The system utilizes a client-server architecture that enables remote profiling, allowing performance data to be captured on a target machine and analyzed on a workstation. It employs lock-free event logging and shared-memory ring buffers to minimize the overhead of data collection, ensuring that the main

Laravel Debugbar is a diagnostic utility and development toolbar designed for the Laravel framework. It functions as an application profiler that monitors runtime performance, memory usage, and database queries to assist in identifying bottlenecks during the development process. The tool integrates directly into the browser, providing a visual interface that displays request data, application state, and performance metrics. By utilizing a collector-based architecture, it aggregates information from various internal framework events and middleware, allowing developers to inspect the applicatio

Booster is an Android app build optimizer and bytecode manipulator designed to reduce binary size, fix system-level crashes, and improve application performance. It functions as an extensible build process plugin that modifies compiled class files and build artifacts to inject stability fixes and optimizations without altering the original source code. The project differentiates itself through low-level bytecode manipulation to patch OS-level bugs and manage thread pools during the compilation phase. It also provides a performance profiling toolkit to identify main-thread blocking operations