使用 Homebrew 安装 apache-spark

项目链接

• 主页
• 仓库
• 上游文档
• 软件包管理器页面

命令和别名

• docker-image-tool.sh
• find-spark-home
• load-spark-env.sh
• pyspark
• run-example
• spark-beeline
• spark-class
• spark-connect-shell
• spark-pipelines
• spark-shell
• spark-sql
• spark-submit
• sparkR

项目历史

Spark originated at the UC Berkeley AMPLab as a faster, more interactive alternative to earlier MapReduce-centered data processing systems. Its project history is closely tied to resilient distributed datasets, in-memory computation, and developer-friendly APIs for Scala, Python, Java, SQL, and R.

Spark became an Apache project and grew into a broad analytics engine rather than a single-purpose batch runner. The official project history notes its Apache Software Foundation path and the release line that made Spark a standard part of the big-data toolchain.

Over time Spark absorbed major adjacent workloads: Spark SQL and DataFrames for structured data, MLlib for machine learning, GraphX for graph processing, Structured Streaming for stream processing, and Spark Connect for client-server connectivity.

采用历史

Spark's adoption history is unusually deep for a package-manager formula because it crossed from research project to de facto data-platform component. It is used for ETL, interactive analytics, machine learning pipelines, and streaming workloads across local machines, YARN, Mesos-era clusters, Kubernetes, and managed cloud services.

The supplied Homebrew package data shows a CLI-heavy install surface: `spark-submit`, `spark-shell`, `pyspark`, `spark-sql`, `sparkR`, `spark-class`, and helper scripts. That executable set mirrors the way Spark became both an application runtime and a command-line toolbox.

使用方式

The main package workflow is submitting applications with `spark-submit`, opening interactive shells with `spark-shell` or `pyspark`, running SQL through `spark-sql`, and configuring behavior through files in `$SPARK_HOME/conf`.

Spark users often install it locally even when production jobs run elsewhere, because the local CLI is useful for testing jobs, validating dependencies, developing notebooks or scripts, and matching cluster runtime behavior.

为什么软件包爱好者会关心

Spark is a classic heavyweight formula: it is mostly scripts plus a large JVM distribution, but those scripts define the ergonomics of a whole data ecosystem. Packagers care about Java compatibility, Python/R bindings, shell wrappers, classpaths, examples, and config file layout.

It is also one of the packages that turns a laptop into a miniature data platform. A formula install can run local mode, submit to clusters, or serve as a client for remote compute, which makes it more than a simple CLI utility.

时间线

• 2009: Spark begins at UC Berkeley AMPLab.
• 2010: Spark is open sourced.
• 2013: Spark enters the Apache Incubator.
• 2014: Spark becomes an Apache top-level project.
• 2020s: Spark continues expanding SQL, streaming, Kubernetes, and client-server features.

Related projects

• Apache Hadoop and YARN are central to Spark's early cluster deployment history.
• Apache Hive influenced Spark SQL's data-warehouse compatibility story.
• Delta Lake, Apache Iceberg, and Apache Hudi are common table-format companions in modern Spark deployments.

来源

• https://github.com/apache/spark
• https://news.apache.org/foundation/entry/the_apache_software_foundation_announces50
• https://spark.apache.org/docs/latest
• https://spark.apache.org/docs/latest/configuration.html
• https://spark.apache.org/history.html
• source_facts.executables