stateright

stateright

Rust实现的分布式系统验证与开发库

Stateright是一个Rust编写的actor库,为分布式算法提供模型检查器、行为探索UI和actor运行时。它内置线性一致性测试器,支持不变量检查、活性检查和对称归约。Stateright允许开发者验证实现的正确性,并可在实际网络上运行系统,无需重新编码。其特点包括全面的测试覆盖和灵活的网络语义选择。

Stateright分布式系统模型检查Actor库RustGithub开源项目

chat crates.io docs.rs LICENSE

Correctly implementing distributed algorithms such as the Paxos and Raft consensus protocols is notoriously difficult due to inherent nondetermism such as message reordering by network devices. Stateright is a Rust actor library that aims to solve this problem by providing an embedded model checker, a UI for exploring system behavior (demo), and a lightweight actor runtime. It also features a linearizability tester that can be run within the model checker for more exhaustive test coverage than similar solutions such as Jepsen.

Stateright Explorer screenshot

Getting Started

  1. Please see the book, "Building Distributed Systems With Stateright."
  2. A video introduction is also available.
  3. Stateright also has detailed API docs.
  4. Consider also joining the Stateright Discord server for Q&A or other feedback.

Examples

Stateright includes a variety of examples, such as a Single Decree Paxos cluster and an abstract two phase commit model.

Passing a check CLI argument causes each example to validate itself using Stateright's model checker:

# Two phase commit with 3 resource managers. cargo run --release --example 2pc check 3 # Paxos cluster with 3 clients. cargo run --release --example paxos check 3 # Single-copy (unreplicated) register with 3 clients. cargo run --release --example single-copy-register check 3 # Linearizable distributed register (ABD algorithm) with 2 clients # assuming ordered channels between actors. cargo run --release --example linearizable-register check 2 ordered

Passing an explore CLI argument causes each example to spin up the Stateright Explorer web UI locally on port 3000, allowing you to browse system behaviors:

cargo run --release --example 2pc explore cargo run --release --example paxos explore cargo run --release --example single-copy-register explore cargo run --release --example linearizable-register explore

Passing a spawn CLI argument to the examples leveraging the actor functionality will cause each to spawn actors using the included runtime, transmitting JSON messages over UDP:

cargo run --release --example paxos spawn cargo run --release --example single-copy-register spawn cargo run --release --example linearizable-register spawn

The bench.sh script runs all the examples with various settings for benchmarking the performance impact of changes to the library.

./bench.sh

Features

Stateright contains a general purpose model checker offering:

  • Invariant checks via "always" properties.
  • Nontriviality checks via "sometimes" properties.
  • Liveness checks via "eventually" properties (experimental/incomplete).
  • A web browser UI for interactively exploring state space.
  • Linearizability and sequential consistency testers.
  • Support for symmetry reduction to reduce state spaces.

Stateright's actor system features include:

  • An actor runtime that can execute actors outside the model checker in the "real world."
  • A model for lossy/lossless duplicating/non-duplicating networks with the ability to capture actor message history to check an actor system against an expected consistency model.
  • Pluggable network semantics for model checking, allowing you to choose between fewer assumptions (e.g. "lossy unordered duplicating") or more assumptions (speeding up model checking; e.g. "lossless ordered").
  • An optional network adapter that provides a lossless non-duplicating ordered virtual channel for messages between a pair of actors.

In contrast with other actor libraries, Stateright enables you to formally verify the correctness of your implementation, and in contrast with model checkers such as TLC for TLA+, systems implemented using Stateright can also be run on a real network without being reimplemented in a different language.

Contribution

Contributions are welcome! Please fork the library, push changes to your fork, and send a pull request. All contributions are shared under an MIT license unless explicitly stated otherwise in the pull request.

License

Stateright is copyright 2018 Jonathan Nadal and other contributors. It is made available under the MIT License.

To avoid the need for a Javascript package manager, the Stateright repository includes code for the following Javascript dependency used by Stateright Explorer:

  • KnockoutJS is copyright 2010 Steven Sanderson, the Knockout.js team, and other contributors. It is made available under the MIT License.

编辑推荐精选

扣子-AI办公

扣子-AI办公

职场AI,就用扣子

AI办公助手,复杂任务高效处理。办公效率低?扣子空间AI助手支持播客生成、PPT制作、网页开发及报告写作,覆盖科研、商业、舆情等领域的专家Agent 7x24小时响应,生活工作无缝切换,提升50%效率!

堆友

堆友

多风格AI绘画神器

堆友平台由阿里巴巴设计团队创建,作为一款AI驱动的设计工具,专为设计师提供一站式增长服务。功能覆盖海量3D素材、AI绘画、实时渲染以及专业抠图,显著提升设计品质和效率。平台不仅提供工具,还是一个促进创意交流和个人发展的空间,界面友好,适合所有级别的设计师和创意工作者。

图像生成AI工具AI反应堆AI工具箱AI绘画GOAI艺术字堆友相机AI图像热门
码上飞

码上飞

零代码AI应用开发平台

零代码AI应用开发平台,用户只需一句话简单描述需求,AI能自动生成小程序、APP或H5网页应用,无需编写代码。

Vora

Vora

免费创建高清无水印Sora视频

Vora是一个免费创建高清无水印Sora视频的AI工具

Refly.AI

Refly.AI

最适合小白的AI自动化工作流平台

无需编码,轻松生成可复用、可变现的AI自动化工作流

酷表ChatExcel

酷表ChatExcel

大模型驱动的Excel数据处理工具

基于大模型交互的表格处理系统,允许用户通过对话方式完成数据整理和可视化分析。系统采用机器学习算法解析用户指令,自动执行排序、公式计算和数据透视等操作,支持多种文件格式导入导出。数据处理响应速度保持在0.8秒以内,支持超过100万行数据的即时分析。

AI工具酷表ChatExcelAI智能客服AI营销产品使用教程
TRAE编程

TRAE编程

AI辅助编程,代码自动修复

Trae是一种自适应的集成开发环境(IDE),通过自动化和多元协作改变开发流程。利用Trae,团队能够更快速、精确地编写和部署代码,从而提高编程效率和项目交付速度。Trae具备上下文感知和代码自动完成功能,是提升开发效率的理想工具。

AI工具TraeAI IDE协作生产力转型热门
AIWritePaper论文写作

AIWritePaper论文写作

AI论文写作指导平台

AIWritePaper论文写作是一站式AI论文写作辅助工具,简化了选题、文献检索至论文撰写的整个过程。通过简单设定,平台可快速生成高质量论文大纲和全文,配合图表、参考文献等一应俱全,同时提供开题报告和答辩PPT等增值服务,保障数据安全,有效提升写作效率和论文质量。

AI辅助写作AI工具AI论文工具论文写作智能生成大纲数据安全AI助手热门
博思AIPPT

博思AIPPT

AI一键生成PPT,就用博思AIPPT!

博思AIPPT,新一代的AI生成PPT平台,支持智能生成PPT、AI美化PPT、文本&链接生成PPT、导入Word/PDF/Markdown文档生成PPT等,内置海量精美PPT模板,涵盖商务、教育、科技等不同风格,同时针对每个页面提供多种版式,一键自适应切换,完美适配各种办公场景。

AI办公办公工具AI工具博思AIPPTAI生成PPT智能排版海量精品模板AI创作热门
潮际好麦

潮际好麦

AI赋能电商视觉革命,一站式智能商拍平台

潮际好麦深耕服装行业,是国内AI试衣效果最好的软件。使用先进AIGC能力为电商卖家批量提供优质的、低成本的商拍图。合作品牌有Shein、Lazada、安踏、百丽等65个国内外头部品牌,以及国内10万+淘宝、天猫、京东等主流平台的品牌商家,为卖家节省将近85%的出图成本,提升约3倍出图效率,让品牌能够快速上架。

下拉加载更多