{"id":13461352,"url":"https://github.com/deepflowio/deepflow","last_synced_at":"2026-01-12T01:55:20.125Z","repository":{"id":37413702,"uuid":"448599559","full_name":"deepflowio/deepflow","owner":"deepflowio","description":"eBPF Observability - Distributed Tracing and Profiling","archived":false,"fork":false,"pushed_at":"2026-01-10T11:48:45.000Z","size":86003,"stargazers_count":3655,"open_issues_count":230,"forks_count":413,"subscribers_count":48,"default_branch":"main","last_synced_at":"2026-01-11T01:02:22.102Z","etag":null,"topics":["apm","gpu","kubernetes","llm","opentelemetry","wasm","zero-code"],"latest_commit_sha":null,"homepage":"https://deepflow.io","language":"Go","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/deepflowio.png","metadata":{"files":{"readme":"README-CN.md","changelog":"CHANGELOG.md","contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":"CODE_OF_CONDUCT.md","threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2022-01-16T15:50:03.000Z","updated_at":"2026-01-10T05:58:37.000Z","dependencies_parsed_at":"2023-12-25T02:42:45.470Z","dependency_job_id":"751f7131-d2b3-469b-b980-19db5f9875eb","html_url":"https://github.com/deepflowio/deepflow","commit_stats":{"total_commits":11158,"total_committers":112,"mean_commits":99.625,"dds":0.8418175300233017,"last_synced_commit":"451006ec0e34b584d8eb7f165901c4ac87b2038b"},"previous_names":["deepflowys/deepflow","deepflowys/metaflow","metaflowys/metaflow"],"tags_count":86,"template":false,"template_full_name":null,"purl":"pkg:github/deepflowio/deepflow","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deepflowio%2Fdeepflow","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deepflowio%2Fdeepflow/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deepflowio%2Fdeepflow/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deepflowio%2Fdeepflow/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/deepflowio","download_url":"https://codeload.github.com/deepflowio/deepflow/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/deepflowio%2Fdeepflow/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":28331314,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-01-12T00:36:25.062Z","status":"ssl_error","status_checked_at":"2026-01-12T00:36:15.229Z","response_time":60,"last_error":"SSL_connect returned=1 errno=0 peeraddr=140.82.121.6:443 state=error: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["apm","gpu","kubernetes","llm","opentelemetry","wasm","zero-code"],"created_at":"2024-07-31T11:00:35.599Z","updated_at":"2026-01-12T01:55:20.120Z","avatar_url":"https://github.com/deepflowio.png","language":"Go","funding_links":[],"categories":["Projects Related to eBPF","10. Application Performance Monitoring Solutions (APM)","Go","Infrastructure","Monitor"],"sub_categories":["Observability","Anomalies Detection"],"readme":"\u003cp align=\"center\"\u003e\n  \u003cimg src=\"./docs/deepflow-logo.png\" alt=\"DeepFlow\" width=\"300\" /\u003e\n\n  \u003cp align=\"center\"\u003eInstant Observability for Cloud \u0026 AI Applications\u003c/p\u003e\n  \u003cp align=\"center\"\u003eZero Code, Full Stack, eBPF \u0026 Wasm\u003c/p\u003e\n\u003c/p\u003e\n\u003cp align=\"center\"\u003e\n    \u003ca href=\"https://zenodo.org/badge/latestdoi/448599559\"\u003e\u003cimg src=\"https://zenodo.org/badge/448599559.svg\" alt=\"DOI\"\u003e\u003c/a\u003e\n    \u003cimg alt=\"GitHub Release\" src=\"https://img.shields.io/github/v/release/deepflowio/deepflow\"\u003e \u003c/a\u003e\n    \u003cimg alt=\"GitCode\" src=\"https://gitcode.com/DeepFlow/deepflow/star/badge.svg\"\u003e \u003c/a\u003e\n    \u003cimg alt=\"docker pulls\" src=\"https://img.shields.io/docker/pulls/deepflowce/deepflow-agent?color=green?label=docker pulls\"\u003e \u003c/a\u003e\n    \u003cimg alt=\"License\" src=\"https://img.shields.io/github/license/deepflowio/deepflow?color=purple\"\u003e \u003c/a\u003e\n\u003c/p\u003e\n\n-------------\n\n简体中文 | [English](./README.md) | [日本語](./README-JP.md)\n\n# 什么是 DeepFlow\n\nDeepFlow 开源项目旨在为复杂的云原生及 AI 应用提供深度可观测性。DeepFlow 基于 eBPF 实现了**零侵扰**（**Zero Code**）的指标、分布式追踪、调用日志、函数剖析数据采集，并结合智能标签（**SmartEncoding**）技术实现了所有观测数据的**全栈**（**Full Stack**）关联和高效存取。使用 DeepFlow，可以让云原生及 AI 应用自动具有深度可观测性，从而消除开发者不断插桩的沉重负担，并为 DevOps/SRE 团队提供从代码到基础设施的监控及诊断能力。\n\n# 核心特性\n\n- **任意** Service 的**全景图**：利用 eBPF **零侵扰**绘制生产环境的全景图，包括任意语言开发的应用服务、AI 服务、基础设施服务。内置标准协议解析能力，并提供 Wasm 插件机制扩展解析任意私有协议。零侵扰计算每一次调用在应用程序和基础设施中的**全栈**黄金指标，快速定界性能瓶颈。\n- **任意** Request 的**分布式追踪**：基于 eBPF 的**零侵扰**分布式追踪能力，支持任意语言的应用程序，并完整覆盖网关、服务网格、数据库、消息队列、DNS、网卡等各类基础设施，不留下任何追踪盲点。**全栈**，自动采集每个 Span 关联的网络性能指标和文件读写事件。从此，分布式追踪进入零插桩的新时代。\n- **任意** Function 的**持续性能剖析**：以低于 1% 的开销**零侵扰**采集生产环境进程的性能剖析数据，绘制函数粒度的 OnCPU、OffCPU、GPU、Memory、Network 火焰图，快速定位业务函数、框架函数、运行时函数、动态链接库函数、内核函数、CUDA 函数的**全栈**性能瓶颈，并自动关联至分布式追踪数据。\n- **无缝集成**流行的可观测性技术栈：可作为 Prometheus、OpenTelemetry、SkyWalking、Pyroscope 的存储后端，亦可提供 **SQL、PromQL、OTLP** 等数据接口作为流行技术栈的数据源。自动为所有观测信号注入统一标签，包括云资源、K8s 容器资源、K8s Label/Annotation、CMDB 中的业务属性等，消除数据孤岛。\n- 存储性能 **10x ClickHouse**：基于 **SmartEncoding** 机制，向所有观测信号注入标准化的、预编码的元标签，相比 ClickHouse 的 String 或 LowCard 方案均可将存储开销降低 10x。自定义标签与观测数据分离存储，从此你可安心注入近乎无限维度和基数的标签，且可获得像 **BigTable** 一样的轻松查询体验。\n\n# 文档\n\n详细信息请访问[文档站点](https://deepflow.io/docs/zh/?from=github)。\n\n# 快速上手\n\nDeepFlow 共有三种版本：\n- DeepFlow Community：DeepFlow 社区版，面向开发人员\n- DeepFlow Enterprise：DeepFlow 企业版，面向组织、解决团队协作的问题\n- DeepFlow Cloud：DeepFlow SaaS 服务，目前处于测试阶段\n\nDeepFlow 社区版由企业版的核心组件构成。通过开源，我们希望让观测更自动，让全世界的开发者更自由。\n\n## 部署 DeepFlow Community\n\n请参考[文档](https://deepflow.io/docs/zh/ce-install/all-in-one/?from=github)部署 DeepFlow Community。\n\n同时我们也搭建了一个完整的 [DeepFlow Community Demo](https://ce-demo.deepflow.yunshan.net/?from=github)，欢迎体验。登录账号 / 密码：deepflow / deepflow-2026\n\n## 体验 DeepFlow Enterprise\n\n你可以访问 [DeepFlow Enterprise Demo](https://deepflow.io/)，目前仅支持中文。\n\n# 从源码编译 DeepFlow\n\n- [编译 deepflow-agent](./agent/build_cn.md)\n\n# 软件架构\n\nDeepFlow Community 版本主要由 Agent 和 Server 两个进程组成。每个 K8s 容器节点、传统服务器或云主机中运行一个 Agent ，负责该服务器上所有应用进程的数据采集。Server 运行在一个 K8s 集群中，提供 Agent 管理、标签注入、数据写入、数据查询服务。\n\n![DeepFlow 软件架构](./docs/deepflow-architecture.png)\n\n# 里程碑\n\n这里有我们[未来的 Feature 规划](https://deepflow.io/docs/zh/about/milestone/?from=github)。欢迎 Issue 和 Pull Request。\n\n# 联系我们\n\n- Discord：点击 [此链接](https://discord.gg/QJ7Dyj4wWM) 加入 Discord 频道.\n- Twitter：[DeepFlow](https://twitter.com/deepflowio)\n- 微信群：\n\u003cimg src=./docs/wechat-group-keeper.png width=30% /\u003e\n\n# 致谢\n\n- 感谢 [eBPF](https://ebpf.io/)，革命性的 Linux 内核技术\n- 感谢 [OpenTelemetry](https://opentelemetry.io/)，提供了采集应用可观测性数据的标准 API\n\n# 荣誉\n\n- DeepFlow 的论文 [Network-Centric Distributed Tracing with DeepFlow: Troubleshooting Your Microservices in Zero Code](https://dl.acm.org/doi/10.1145/3603269.3604823) 已被国际顶会 ACM SIGCOMM 2023 录用\n- DeepFlow 已加入 \u003ca href=\"https://landscape.cncf.io/?selected=deep-flow\"\u003eCNCF CLOUD NATIVE Landscape\u003c/a\u003e\n- DeepFlow 已加入 \u003ca href=\"https://landscape.cncf.io/?selected=deep-flow\u0026group=cnai\u0026item=cnai--model-llm-observability--deepflow\"\u003eCNCF CNAI (Cloud-Native AI) Landscape\u003c/a\u003e\n- DeepFlow 已加入 \u003ca href=\"https://ebpf.io/applications#deepflow\"\u003eeBPF Project Landscape\u003c/a\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeepflowio%2Fdeepflow","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fdeepflowio%2Fdeepflow","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fdeepflowio%2Fdeepflow/lists"}