Cloud Parts

The Apache Software Foundation

The non-profit behind Apache Kafka, Spark, Iceberg and about 300 other projects, and the author of the Apache License 2.0 that many non-Apache projects use too.

38 projects in this catalogWebsite

What it is

The Apache Software Foundation (ASF) is a US 501(c)(3) charity, incorporated in 1999, that holds the legal side of a few hundred open-source projects: their trademarks, the right to distribute their code, the servers they run on, and the money from donations and sponsors. It doesn’t write the software. Each project is run by its own volunteers, and the foundation pays nobody, not even its members or officers, to work on it; many committers are paid by their employers, but never by the ASF. In its 2026 fiscal year it counted 10,225 committers across 302 projects.

“Apache” in a name, as in Apache Kafka or Apache Spark, means the project lives at the ASF: the name is an ASF trademark, the project’s own committee votes its releases out under the Apache License 2.0, and decisions are made in public on its mailing lists. It doesn’t mean the projects share code, or that they have anything to do with the Apache HTTP Server, the web server the foundation grew out of. “Apache 2.0” on its own means only the licence, which plenty of projects outside the ASF use, Kubernetes among them.

For someone choosing a technology, the foundation is a promise about governance more than about code. Contributors keep the copyright to their work and license it to the ASF; the foundation requires every release to be under the Apache License 2.0 and won’t dual-license; and no company, however many committers it employs or however much it sponsors, gets special privileges or control over a project’s direction. So no single company can take an Apache project source-available, as happened to Elasticsearch, Redis, Terraform and Akka, which were each controlled by one company. A company that employs most of a project’s committers can still walk away, and anyone may build a closed product on the code, but the Apache-named project, its trademark and its licence stay with the foundation.

Its projects in this catalog

Compute

Apache PekkoOpen sourceThe Apache-licensed fork of Akka 2.6: actors, cluster sharding, persistence, streams and HTTP for Java and Scala, run by a foundation instead of a company.

Networking

Apache ThriftOpen sourceFacebook’s 2007 RPC framework: define types and services in one IDL file, generate clients and servers for over 20 languages, and pick the wire protocol and transport separately.

Messaging

Apache FlinkOpen sourceThe most widely used open-source engine for stateful stream processing: keyed state, event time and exactly-once recovery from checkpoints.Apache FlussOpen sourceStreaming storage built for Flink: event logs and updatable tables in columnar Arrow format, with key lookups, and older data copied into Paimon, Iceberg, Hudi or Lance tables.Apache KafkaOpen sourceThe open-source distributed log that most event streaming is built on, or built to be compatible with.Apache PulsarOpen sourceA multi-tenant messaging and streaming platform from Yahoo that keeps brokers stateless and stores messages in Apache BookKeeper, so one topic can serve as a log, a pub-sub feed or a work queue.Kafka StreamsOpen sourceA Java library, part of Apache Kafka, that turns your own application into a stateful stream processor with no cluster of its own.

Storage

Apache OpenDALOpen sourceA Rust library, with bindings for Python, Java, Node.js, Go and more, that gives one API (read, write, stat, list, delete) over 50-plus storage backends: S3 and other object stores, local and HDFS file systems, Google Drive and Dropbox, and key-value stores.Apache OzoneOpen sourceA Hadoop-born object store that serves the same buckets over S3 and as a Hadoop file system.

Databases

Apache CassandraOpen sourceA masterless, wide-column database that spreads and replicates partitions around a ring of equal nodes, with consistency chosen per request.Apache HBaseOpen sourceThe Hadoop world’s Bigtable: huge sorted tables split into regions, each served by one RegionServer, with files kept in HDFS.Apache IoTDBOpen sourceAn Apache time-series database for industrial IoT that stores everything in its own open file format, TsFile, and lets you model devices as a tree of paths or as SQL tables.Apache ZooKeeperOpen sourceThe coordination service from Yahoo that a generation of Hadoop-era systems use for leader election, membership and configuration.

Analytics

Apache AmoroIncubatingOpen sourceAn incubating Apache “lakehouse management system”, started at NetEase as Arctic, that watches Iceberg tables and keeps them fast on its own: it compacts small files, rewrites deletes and expires old snapshots and files in the background.Apache ArrowOpen sourceA standard way to lay out columnar data in memory, so libraries, engines and languages can share tables without copying or converting them.Apache Arrow AceroExperimentalOpen sourceThe streaming execution engine inside Arrow’s C++ library: you hand it an exact plan of nodes (scan, filter, project, aggregate, join, sort) and it streams Arrow record batches through them, with no SQL parser or optimizer of its own.Apache AvroOpen sourceA compact, row-based binary format with a JSON schema, widely used for Kafka messages and record files.Apache CalciteOpen sourceA Java framework with everything a database needs to understand a query (SQL parser, validator, relational algebra and a rule-based, cost-based optimizer) but no storage of its own, which Hive, Flink, Druid, Dremio, Beam and many other systems embed as their SQL front end and planner.Apache DataFusionOpen sourceA query engine library in Rust, built on Arrow, that other projects embed and extend to build their own databases and data tools.Apache DataFusion CometOpen sourceA Spark plugin that keeps Spark’s optimizer and scheduler but runs the physical plan on Apache DataFusion in Rust, over Arrow batches, with no changes to your Spark code, and falls back to Spark for anything it can’t run the same way.Apache DorisOpen sourceAn open-source MPP analytical database, born at Baidu as Palo, that serves sub-second queries on fresh, updatable data through the MySQL protocol and also queries Iceberg, Hudi, Paimon and Hive tables in place.Apache DruidOpen sourceAn open-source real-time analytics database that makes Kafka and Kinesis events queryable within seconds and serves fast slice-and-dice queries to dashboards and APIs.Apache HadoopOpen sourceThe project that made big-data clusters normal: a distributed file system (HDFS), a cluster scheduler (YARN) and the MapReduce engine.Apache HiveOpen sourceThe first SQL engine for Hadoop, and the source of the Hive Metastore, the table catalog that Trino, Spark, Impala and Flink still read.Apache HudiOpen sourceA table format built for upserts and incremental pulls, with record-level indexes and built-in table services.Apache IcebergOpen sourceThe most widely supported open table format: snapshots of data files, tracked by a tree of metadata and swapped in atomically through a catalog.Apache ImpalaOpen sourceAn MPP SQL engine in C++ for interactive queries over Hive tables on HDFS, Kudu, Iceberg and object storage; built and mainly sold by Cloudera.Apache NiFiOpen sourceA Java dataflow server you program by drawing a graph of processors in a web UI, with buffered queues, back pressure and a record of where every piece of data went.Apache ORCOpen sourceA columnar file format born in Hive: typed columns in large stripes, built-in indexes every 10,000 rows, and the storage behind Hive’s ACID tables.Apache PaimonOpen sourceA lake table format built on LSM trees, so streaming jobs can upsert into object storage and other jobs can read the changes as a stream.Apache ParquetOpen sourceThe standard columnar file format for analytics: each column stored and compressed on its own, with statistics that let engines skip data.Apache PigDormantOpen sourceA step-by-step scripting language, Pig Latin, that compiles to MapReduce, Tez or Spark jobs on Hadoop. Still released, but the project calls itself dormant.Apache PolarisOpen sourceAn open-source Iceberg REST catalog: one server that every engine asks where a table’s current metadata is, with role-based access control and short-lived storage credentials per table.Apache SparkOpen sourceThe general-purpose distributed engine for large batch jobs, streaming and machine learning, in Python, SQL, Scala, Java or R.Apache TsFileOpen sourceA columnar file format for device time series, with per-series indexes and time-series encodings; the storage format of Apache IoTDB and its own Apache project since 2023.Apache XTableIncubatingOpen sourceAn incubating Apache project, formerly Onehouse’s OneTable, that makes one set of Parquet files readable as a Delta Lake, Apache Iceberg and Apache Hudi table at once by translating only the table metadata.Apache ZeppelinOpen sourceA self-hosted web notebook where each paragraph runs through an interpreter: Spark, Flink, JDBC, Python and more. Maintained, but behind Jupyter and Databricks in reach.PuffinOpen sourceApache Iceberg’s file format for the data that doesn’t fit in a manifest: a container of typed binary blobs, such as distinct-value sketches and deletion vectors, described by a JSON footer.

In depth

How a project is run#

The board sets up a Project Management Committee (PMC) for every top-level project. Contributors send patches, documentation, bug reports and answers on the mailing lists. When the project feels someone has earned it, it makes them a committer: write access to the repository, an apache.org address, and a signed Contributor License Agreement on file. PMC members are committers elected to the committee: they cast the binding votes, propose new committers, and must vote to approve every release. The board appoints one PMC member as chair, who becomes a Vice President of the foundation and is the link to the board. ASF members, elected by existing members, are legally the foundation’s owners, like shareholders, and elect the nine-person board every year.

The foundation calls its culture “The Apache Way” and says there’s no single definition of it. The parts it lists are earned authority (influence comes from what you contribute, belongs to you rather than your employer, and doesn’t carry over to another project), a community of peers in which individuals take part, not companies, and “benevolent dictators for life” are not allowed, open communication, consensus decisions and independence from any vendor. Work happens on public, archived mailing lists such as dev@ and user@; decisions about code or direction made in private have to be brought back on-list. As one foundation report puts it, “If it didn’t happen on-list, it didn’t happen.”

Most decisions use lazy consensus: a proposal goes ahead with a few +1 votes and no -1, and anyone who votes -1 must explain why or offer an alternative. On a code change, a -1 is a veto the PMC can’t override. Releases follow their own rule: at least three binding +1 votes from PMC members and more +1 than -1 votes, and a release can’t be vetoed. Each PMC reports the project’s health to the board every quarter (monthly for the first three months after leaving the Incubator), and the board can dissolve a PMC at any time.

From Incubator to Attic#

Projects don’t start at the top level. In fiscal 2026 (May 2025 to April 2026) the foundation reported 302 projects run by 210 committees, 10,225 committers, 1,188 members and 1,709 releases; 10 new podlings entered the Incubator, 11 projects graduated and 7 were retired.

Incubator

28 podlings in FY2026

A project that wants to join brings a proposal, a working codebase, the intention to grant the ASF rights to its code, and a sponsoring ASF member. It becomes a “podling”, with mentors who teach it the Apache Way, and the Incubator reports on it to the board every month. The Incubator judges community, not technology: the main test is whether the podling grows a diverse group of committers, not all from one employer. Incubation typically takes about a year and a half; by April 2023, 239 of 335 podlings had graduated. Kafka entered in 2011 and Spark in 2013.

Top-level project

11 graduated in FY2026

When the Incubator recommends it, the board graduates the podling into a top-level project with its own PMC, which then releases on its own authority and reports straight to the board. A top-level project has full say over its technical direction; the board only checks that the community is healthy and the legal process is followed.

The Attic

7 retired in FY2026

A project that can’t muster three votes for a release, has no active committers or stops reporting to the board is a candidate for the Attic, set up in November 2008. The move happens by the PMC’s own choice or a board resolution. The Attic keeps the source code, documentation and records available but makes no fixes or releases. Retired projects include Apache Sqoop (June 2021), Apache Oozie (February 2025) and Apache Mesos (August 2025). A retired project can be restarted in the Incubator, handed to another PMC, or forked outside the ASF.

The Apache License 2.0#

The foundation wrote the Apache License; version 2.0 dates from January 2004 and is the licence every ASF project releases under. It lets anyone use, change and redistribute the code, as source or binaries, in open-source or closed products, with no fee. Each contributor also grants a patent licence covering their contributions, which ends for anyone who sues claiming the work infringes a patent.

The conditions are about notices: ship a copy of the licence, mark the files you changed, keep the existing copyright notices, and if the project has a NOTICE file, pass its attribution notices along. The licence gives no right to use the project’s name or trademarks. You may put your own changes, or a whole derived work, under different terms, which is why companies can sell closed products and services built on Apache code.

Compatibility with the GPL runs one way. The Free Software Foundation counts Apache 2.0 as compatible with GPLv3, so Apache-licensed code can go into GPLv3 projects, but the ASF won’t take GPL code into its own projects. The FSF says Apache 2.0 is not compatible with GPLv2, because of the patent termination and indemnification requirements GPLv2 doesn’t have.

The licence alone doesn’t stop a change of licence: Elasticsearch, Akka and MinIO were all Apache 2.0 until the companies that controlled them moved new releases to other terms. What stops it at the ASF is who holds the rights. Contributors license their work to the foundation, the foundation requires every release to be under Apache 2.0 and refuses to dual-license, and its projects may not distribute under anything more restrictive. Releases already out keep their licence whatever happens, which is what forks start from.

When one company holds the rights#

Projects on this site whose owners changed the licence of new releases. None of them was an ASF project; the forks started from the last release under the old licence.

ProjectLicence beforeLicence afterBest-known fork
Elasticsearch (Elastic, January 2021)Apache 2.0SSPL or Elastic License from 7.11; AGPLv3 added as a third option in August 2024OpenSearch, from Elasticsearch 7.10.2, Apache 2.0, a Linux Foundation project
Akka (Lightbend, September 2022)Apache 2.0Business Source License 1.1: production use needs a commercial licence, free under US$25 million revenue; each version reverts to Apache 2.0 after three yearsApache Pekko, from Akka 2.6.x, now an ASF project
Terraform (HashiCorp, August 2023)MPL 2.0Business Source License 1.1 for all future releasesOpenTofu, given to the Linux Foundation
Redis (Redis Ltd., March 2024)BSD 3-clauseRSALv2 or SSPLv1 from 7.4; AGPLv3 added as an option with Redis 8 in May 2025Valkey, from Redis 7.2.4, BSD 3-clause, a Linux Foundation project
MinIO (MinIO, Inc., April 2021)Apache 2.0AGPLv3 (still open source); the open-source repository has since been archivedNot covered here
  • The vendors themselves call SSPL, RSAL and the BSL “source-available” licences. AGPLv3 is an OSI-approved open-source licence.

“Apache” is not one thing#

The Apache HTTP Server

the original project

The web server the foundation grew out of, often simply called “Apache”. Its first public release, 0.6.2, came out in April 1995, and within a year of 1.0 it was the most-used web server. Today it is one ASF project among hundreds; Kafka or Spark share nothing with it but the foundation.

The Apache License outside the ASF

licence only

Anyone can use Apache 2.0. Kubernetes is Apache 2.0 but belongs to the Cloud Native Computing Foundation, whose charter makes Apache 2.0 the licence for all its code; OpenSearch is Apache 2.0 under the Linux Foundation. “Apache 2.0” on a project page tells you the licence, not who governs the project.

Linux Foundation and CNCF

501(c)(6)

The Linux Foundation is a 501(c)(6) non-profit rather than a 501(c)(3) charity like the ASF. CNCF is part of it and hosts Kubernetes, Prometheus and Envoy. Valkey and OpenTofu, the forks made after the Redis and Terraform licence changes, also went to the Linux Foundation.

Eclipse Foundation

Brussels

Set up in January 2004 as an independent not-for-profit to steward the Eclipse community, now based in Brussels, with Jakarta EE and more than 450 other projects. Its projects are also grouped under PMCs, whose leads are appointed by the Eclipse Foundation’s board, much as the ASF board appoints PMC chairs.

Where it came from1995–2025

In early 1995 work on NCSA HTTPd, the web server written at NCSA for users of the Mosaic browser, had stalled after its author left. Brian Behlendorf started collecting patches against the last NCSA release, and a group of eight, the Apache Group, traded them on a mailing list. Apache 0.6.2 came out in April 1995 and 1.0 on December 1, and within a year it had passed NCSA as the most-used web server. The ASF says Behlendorf chose the name to show appreciation for the Apache peoples; the “a patchy server” pun spread later, and in the foundation’s words “rumor became lore, and lore became legend.”

As other projects grew up around the server, the group incorporated the foundation as a non-profit on June 1, 1999, so the projects would outlive any individual volunteer. The Incubator followed in 2002 and the Apache License 2.0 in 2004. Lucene graduated in 2005, Hadoop in 2008, and HBase, Hive, Pig, Cassandra and ZooKeeper in 2010, making the ASF home to much of the big-data stack.

In September 2025 the foundation replaced its feather logo with an oak leaf, in a process guided by a Logo PMC of members, and began using “The ASF” in its visual identity. It kept its legal name and every project name, citing the supply-chain risk of moving away from apache.org. In December 2025 it welcomed its 10,000th committer.

Why it’s built this way

A charity that holds the rights, not a company
Incorporating in 1999 gave the projects a legal owner that would outlive individual volunteers, could take donations, shield volunteers from lawsuits aimed at the projects, and protect the Apache name from abuse.
Individuals take part, not companies
Merit belongs to the person and doesn’t move with their employer, and no sponsor or employer gets special privileges, so no single vendor can steer a project.
Projects govern themselves; the board oversees
By 1999 there were several communities with their own habits. Rather than force one set of rules from the top, the ASF made each PMC the decision maker for its project, with the board checking process and community health through reports.
Incubate for community, not technology
A project whose committers all work for one employer can collapse when that employer leaves, so the Incubator judges whether a podling can grow a diverse, self-governing community rather than whether its technology is best.

Timeline

  1. 1995

    The eight-person Apache Group patches NCSA HTTPd; Apache 1.0 ships December 1

  2. 1999

    The ASF incorporates on June 1 as a 501(c)(3) non-profit

  3. 2002

    The board creates the Apache Incubator in October

  4. 2004

    Apache License 2.0 published in January

  5. 2008

    Hadoop becomes a top-level project; the Attic is set up in November

  6. 2010

    HBase, Hive, Pig, Cassandra and ZooKeeper become top-level projects

  7. 2025

    Oak-leaf logo replaces the feather in September; 10,000th committer in December

Sources · facts checked 2026-10-06

Improve this page on GitHub