# apache/kafka

Apache Kafka - A distributed event streaming platform

Repository: https://github.com/apache/kafka
Canonical: https://ross.abutalabs.com/products/kafka
Homepage: https://kafka.apache.org/
Language: Java
License: Apache-2.0
License Family: permissive
Topics: scala, kafka, java, streaming
Last push: 2026-08-26T21:27:56+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 35, longevity 100
- inputs: {"age_days": 5497, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 33632, forks 15461 (observed 2026-08-28T04:11:58.382009+00:00)

## What it is
Apache Kafka is an open-source distributed event streaming platform for building high-performance data pipelines, streaming analytics, and data integration. It provides durable, fault-tolerant storage of event streams with built-in stream processing (Kafka Streams) and a Connect framework for integrating with external systems.

## Use cases
- build high-throughput data pipelines between systems
- stream and process events in real time
- decouple microservices with a publish-subscribe event bus
- ingest and store application logs, metrics, and sensor data durably
- integrate event sources and sinks like Postgres, Elasticsearch, and S3
- run exactly-once stream processing with joins and aggregations
- handle trillions of messages per day across a scalable cluster

## When to choose
- you need durable, ordered, fault-tolerant event storage at massive scale
- you want battle-tested infrastructure trusted by thousands of large organizations
- you need built-in stream processing and a wide ecosystem of connectors
- you require guaranteed ordering, zero message loss, or exactly-once semantics

## When to avoid
- you only need simple point-to-point messaging or a lightweight task queue
- your team cannot operate a JVM-based distributed cluster
- your message volume is small and a database or simpler broker would suffice
- you need per-message complex querying rather than append-only streams

## Facets
- artifact type: service
- maturity: stable
- function: message-queue, streaming, etl, data-science
- domain: microservices, big-data, analytics
- platform: jvm, windows, cloud, self-hosted
- tags: event-streaming, pub-sub, log-storage, stream-processing, kafka-connect, kafka-streams, exactly-once, high-throughput, data-engineering, messaging, real-time, docker, linux, macos

## Member repositories
- apache/kafka (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:58.382009+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:28:13.624724+00:00, confidence not recorded.
  - readme: https://github.com/apache/kafka (fetched 2026-08-28T04:11:58.382009+00:00, sha 351690219ec5)
  - homepage: https://kafka.apache.org/ (fetched 2026-08-29T07:48:35.985942+00:00, sha 97856c12493c)
  - site_page: https://kafka.apache.org/43/getting-started (fetched 2026-08-29T07:48:35.988748+00:00, sha 716c4e4bb65b)
  - site_page: https://kafka.apache.org/quickstart (fetched 2026-08-29T07:48:35.990597+00:00, sha 4d9fb093039d)
  - site_page: https://kafka.apache.org/documentation (fetched 2026-08-29T07:48:35.992741+00:00, sha e25789db26ce)
  - site_page: https://kafka.apache.org/documentation/streams (fetched 2026-08-29T07:48:35.995673+00:00, sha a91d5760408c)
- Data as of 2026-08-30T08:39:29.467469+00:00.