apache/datafusion-ballista：README 来源编辑指南

apache/datafusion-ballista 的 README 将项目描述为"Apache DataFusion Ballista Distributed Query Engine"。本文只整理仓库能直接核验的内容，不把星标、Fork 或宣传语当成质量证明。README 在"README"下的说明是：<!--- Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership.。这给出的首先是项目边界，而不是已经完成的生产验证。

从 README 的"Who is Ballista for"和相关条目看，读者可以先判断它是否解决自己的具体问题：Spark users wanting the same execution model , you run Spark SQL or batch jobs and want a lighter, Rust-native alternative without relearning a new paradigm.。如果你的目标与这段说明不一致，就不应仅凭项目热度采用它。这里保留原项目名、命令和组件名，方便回到一手来源核对。 README 还列出了另一条可核对的信息：DataFusion users going multi-node , you already use Apache DataFusion on a single machine and have outgrown it. Ballista runs the same SQL and DataFrame workloads across a cluster with minimal code changes and the same results.。这类原文条目可以帮助读者设计试运行步骤，但不能代替自己的环境测试。

README 把工作方式分散写在"Ballista: Making DataFusion Applications Distributed"等段落中。可确认的线索包括：Ballista is a distributed query execution engine that enhances Apache DataFusion by enabling the parallelized execution of workloads across multiple nodes in a distributed environment.。这篇整理没有把未写出的架构、性能或安全边界补成结论；真正的运行链仍应结合仓库目录、配置文件和版本标签检查。

第一次安装应从 README 给出的入口开始。当前可复核的命令是： # Build with standalone support (default) cargo build -p ballista # Build with Substrait support cargo build -p ballista-scheduler --features substrait # Build with Spark compatibility cargo build -p ballista-executor --features spark-compat 如果仓库没有提供命令，本文不会替它编造安装步骤，而是建议先打开 README 的"Who is Ballista for"部分，确认系统依赖、默认端口和首次初始化动作。

日常使用的细节取决于项目实际文档。README 的"Ballista: Making DataFusion Applications Distributed"段落提到：can be distributed with few lines of code changed:。对于配置文件、环境变量、权限和数据目录，当前稿只记录来源明确的部分；未写明的默认值必须在测试环境中验证，并保留可回滚的配置副本。 同一部分还提到：Library users building a specialized engine , you are building a bespoke distributed query engine and want reusable scheduler, executor, and plan-serialization building blocks with extension points, instead of writing distributed execution。