登录

dgraph

dgraph-ioApache-2.0活跃维护

高性能分布式图数据库，专为实时查询场景设计，支持知识图谱等大规模关联数据的存储与毫秒级检索。

high-performance graph database for real-time use cases

v25.5.1发布于 2026-10-22 15:56 UTC21,823star +3 / 7 天1,610fork同步于 2026-11-10 05:22 UTC
图数据库分布式实时查询知识图谱Go
对比 GitHub 官网databasedistributedgoknowledge-graph

怎么上手

由 README 提炼，执行前请确认
原文 
docker pull dgraph/dgraph:latest
docker run -it -p 8080:8080 -p 9080:9080 -v ~/dgraph:/dgraph dgraph/standalone:latest
硬件要求按 README 说明

README 未说明具体显存或内存要求；官方仅支持 Linux/amd64 与 Linux/arm64，2021 年起不再官方支持 Mac 和 Windows，推荐用 Docker 部署。

有 Docker 镜像
下载：源码 zip·GitHub Releases（安装包）本站不镜像安装包

项目说明

由 README 自动翻译提炼
没人工点评？写下你的经验

原生 GraphQL 的分布式图数据库，用图后端支撑实时查询与海量结构化数据

高性能分布式图数据库，专为实时查询场景设计，支持知识图谱等大规模关联数据的存储与毫秒级检索。

核心能力

  • 提供分布式 ACID 事务、一致性复制与线性化读
  • 原生支持 GraphQL 查询语法，返回 JSON 或 Protobuf
  • 通过 GRPC 与 HTTP 对外提供接口
  • 分片分布式架构，支持自动数据迁移与再平衡
  • 原生支持全文检索、正则与地理搜索

容易踩的坑

从源码构建需 Go 1.27+ 及 build-essential 等依赖；官方已不支持 Mac/Windows，其他平台可编译但无官方支持；standalone 容器需挂载卷并映射 8080/9080 端口才能持久化与访问。

内容由模型根据 README 生成，可能不完整或不准确，以官方文档为准。看原文 README

最近发版

5 个版本v25.5.1
全部 release
  1. v25.5.1最新正式版2026-10-22 · 2026-10-22 15:56 UTC

    What's Changed

    New Contributors

    Full Changelog: https://github.com/dgraph-io/dgraph/compare/v25.5.0...v25.5.1

  2. v25.5.02026-10-13 · 2026-10-13 19:37 UTC
    查看更新说明

    What's Changed

    New Contributors

    Full Changelog: https://github.com/dgraph-io/dgraph/compare/v25.4.1...v25.5.0

  3. v25.4.12026-08-24 · 2026-08-24 19:28 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/dgraph-io/dgraph/compare/v25.4.0...v25.4.1

  4. v25.4.02026-07-30 · 2026-07-30 18:23 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/dgraph-io/dgraph/compare/v25.3.8...v25.4.0

  5. v25.3.82026-07-09 · 2026-07-09 00:31 UTC
    查看更新说明

    What's Changed

    Full Changelog: https://github.com/dgraph-io/dgraph/compare/v25.3.7...v25.3.8

用户点评

登录用户的真实使用记录，可按版本和硬件筛选
共 0 条

    还没有人点评。第一个说出真实体验的人最有价值。

    README 原文

    未翻译
    在 GitHub 打开

    GitHub Repo stars GitHub commit activity Go Report Card Docker Pulls

    Dgraph is a horizontally scalable and distributed GraphQL database with a graph backend. It provides ACID transactions, consistent replication, and linearizable reads. It's built from the ground up to perform a rich set of queries. Being a native GraphQL database, it tightly controls how the data is arranged on disk to optimize for query performance and throughput, reducing disk seeks and network calls in a cluster.

    Dgraph's goal is to provide Google production-level scale and throughput, with low enough latency to serve real-time user queries over terabytes of structured data. Dgraph supports GraphQL query syntax, and responds in JSON and Protocol Buffers over GRPC and HTTP. Dgraph is written using the Go Programming Language.

    Status

    Dgraph is at version v25 and is production-ready. Apart from the vast open source community, it is being used in production at multiple Fortune 500 companies.

    Supported Platforms

    Dgraph officially supports the Linux/amd64 and Linux/arm64 architectures. In order to take advantage of memory performance gains and other architecture-specific advancements in Linux, we dropped official support for Mac and Windows in 2021, see this blog post for more information. You can still build and use Dgraph on other platforms (for live or bulk loading for instance), but support for platforms other than Linux/amd64 and Linux/arm64 is not available.

    Running Dgraph in a Docker environment is the recommended testing and deployment method.

    Install with Docker

    If you're using Docker, you can use the official Dgraph image.

    docker pull dgraph/dgraph:latest

    For more information on a variety Docker deployment methods including Docker Compose and Kubernetes, see the docs.

    Run a Quick Standalone Cluster

    docker run -it -p 8080:8080 -p 9080:9080 -v ~/dgraph:/dgraph dgraph/standalone:latest

    Install from Source

    If you want to install from source, install Go 1.27+ or later and the following dependencies:

    Ubuntu

    sudo apt-get update
sudo apt-get install build-essential

    Build and Install

    Then clone the Dgraph repository and use make install to install the Dgraph binary in the directory named by the GOBIN environment variable, which defaults to $GOPATH/bin or $HOME/go/bin if the GOPATH environment variable is not set.

    git clone https://github.com/dgraph-io/dgraph.git
cd dgraph
make setup
make install

    Get Started

    To get started with Dgraph, follow:

    Is Dgraph the right choice for me?

    • Do you have more than 10 SQL tables connected via foreign keys?
    • Do you have sparse data, which doesn't elegantly fit into SQL tables?
    • Do you want a simple and flexible schema, which is readable and maintainable over time?
    • Do you care about speed and performance at scale?

    If the answers to the above are YES, then Dgraph would be a great fit for your application. Dgraph provides NoSQL like scalability while providing SQL like transactions and the ability to select, filter, and aggregate data points. It combines that with distributed joins, traversals, and graph operations, which makes it easy to build applications with it.

    Dgraph compared to other graph DBs

    FeaturesDgraphNeo4jJanus Graph
    ArchitectureSharded and DistributedSingle server (+ replicas in enterprise)Layer on top of other distributed DBs
    ReplicationConsistentNone in community edition (only available in enterprise)Via underlying DB
    Data movement for shard rebalancingAutomaticNot applicable (all data lies on each server)Via underlying DB
    LanguageGraphQL inspiredCypherGremlin
    ProtocolsGrpc / HTTP + JSON / RDFBolt + CypherWebsocket / HTTP
    TransactionsDistributed ACID transactionsSingle server ACID transactionsNot typically ACID
    Full-Text SearchNative supportNative supportVia External Indexing System
    Regular ExpressionsNative supportNative supportVia External Indexing System
    Geo SearchNative supportExternal support onlyVia External Indexing System
    LicenseApache 2.0GPL v3Apache 2.0

    Users

    Developers

    Please see Contributing to Dgraph for guidelines on contributions.

    Client Libraries

    The Dgraph team maintains several officially supported client libraries. There are also libraries contributed by the community unofficial client libraries.

    Contact

    相关项目

    同分类里 star 最高的项目
    对比全部 →
    LocalAI
    mudler·Go

    开源本地 AI 推理引擎，无需 GPU 即可在任意硬件上跑 LLM、视觉、语音和图像生成模型，提供兼容 OpenAI 的 API。

    推理引擎+4分布式训练
    50281+454.6kv4.12.12026-11-10 01:47 UTC
    DeepSpeed
    deepspeedai·Python

    基于 PyTorch 的深度学习优化库，通过 ZeRO、模型并行和混合专家等技术，让千亿到万亿参数模型的分布式训练与推理更省显存、更高效。

    分布式训练+4分布式训练
    43315+85.1kv0.19.92026-11-10 01:13 UTC
    ColossalAI
    hpcaitech·Python

    一套面向大模型训练的分布式并行框架，把数据、模型、流水线并行和异构训练整合成统一接口，让训练和推理更省显存、更快。

    分布式训练+4分布式训练
    41431+34.5kv0.5.02026-11-10 04:53 UTC
    horovod
    horovod·Python

    Distributed training framework for TensorFlow, Keras, PyTorch, and Apache MXNet.

    分布式训练
    146742.2k2026-09-17 09:15 UTC