Apache Fluss™ Skip to main content DocsBlogLearnTalks Videos CommunityRoadmapDownloads Ask AI Search Open Source · Apache 2.0Streaming Storage for Real-Time Analytics & AI Apache Fluss is an open-source, lakehouse-native streaming storage. It collapses the message broker, online KV store, stream-processing state backend, and lakehouse cold store into a single coherent foundation, making the Lakehouse truly real-time. Get Started→View on GitHub Flink SQLSpark SQL -- Register Apache Fluss as a Flink catalog CREATE CATALOG fluss_catalog WITH ( 'type' = 'fluss', 'bootstrap.servers' = 'coordinator-server:9123' ); USE CATALOG fluss_catalog; -- Create a primary-key table CREATE TABLE pk_table ( shop_id BIGINT, user_id BIGINT, num_orders INT, PRIMARY KEY (shop_id, user_id) NOT ENFORCED ) WITH ('bucket.num' = '4'); INSERT INTO pk_table VALUES (1234, 1234, 1); SELECT * FROM pk_table WHERE shop_id = 1234; ArchitectureUnlocking the Lakestream Architecture The multiple-systems taxFive systems, four integrations, continuous engineering tax. A conventional real-time AI stack stitches together a message broker, a stream processor, an online store, an offline store, and a synchronization layer. Every boundary is an integration point where data silently diverges. Apache Fluss collapses that stack into one substrate. Before · fragmented stack01Message broker Kafka, for event transport. 02Stream processor Flink or Spark, for derived features and aggregations. 03Online store Redis or DynamoDB, for sub-millisecond lookup. 04Offline store Iceberg or Parquet on S3, for training and history. 05Sync layer Bespoke pipelines and freshness monitors that drift silently. 5 Systems · 4 Sync Boundaries · Continuous Engineering Tax After · unified substrateApache Fluss One columnar streaming store designed for the real-time AI data plane. ✓Streaming Log · Durable, replayable, offset-ordered streams ✓PK Lookup · Sub-millisecond key/value serving ✓Lakestream · Real-time data layer for Lakehouse architecture ✓State Store · Externalized state for joins and aggregations ✓Multi-Modal · Lance integration for vectors and ML context ✓Audit Trail · Change data feed, replayable by design 1 Substrate · 0 Sync Boundaries · Single Source of Truth Six capability pillarsThe benefits, grounded in the architecture. Each pillar is a direct consequence of a specific architectural mechanism. Together they collapse the fragmented real-time stack into a single coherent foundation. 01 Unified Architecture One system for messaging, applications, analytics, and AI. Replaces the message queue, key-value store, and OLAP engine with a single platform serving transport, lookups, and queries from the same data. Architectural basisDual representation of PK Tables (Log Store & KV Store). 02 Stream & Lakehouse Unification One copy of data across real-time and batch layers. Hot and cold tiers share the same schema and are queryable as one substrate, so streaming and historical reads hit one source of truth. Architectural basisTiering Service and Union Read across Iceberg, Paimon, and Lance. 03 Compute / Storage Separation Lean, elastic, stateless compute with fast recovery. Stateless compute recovers in seconds and runs up to 85% cheaper than Kafka-based topologies. State lives on the Fluss leader, not Flink slots. Architectural basisStateless compute model with leader-resident state and KV snapshots. 04 Columnar Streaming Analytics Pruning that compounds. Server-side projection, predicate pushdown, and partition pruning on Arrow-format streams compound into order-of-magnitude I/O and network savings. Architectural basisARROW log format and the compound pruning stack on the TabletServer. 05 Feature & Context Stores Multi-modal data on one substrate, ready for ML and AI. Row, columnar, and vector data on one store. Online features, RAG context, and analytics collapse into one PK Table accessed through different views. Architectural basisUnified substrate spanning structured features and vector context. 06 Ecosystem Openness Open formats. No vendor lock-in. Readable by Flink, Spark, Trino, StarRocks, Doris, and DuckDB. Native hot tier plus Iceberg, Paimon, and Lance for the cold tier, open formats end to end. Architectural basisOpen lake formats throughout, governed at the Apache Software Foundation. Apache Fluss vs Apache KafkaWhere Streams Meet The Lakehouse Kafka is the streaming transport. Fluss is the streaming storage. If your need is large-scale stream processing with Flink, real-time analytics, AI/ML, or a sub-second lakehouse, Fluss is the shared streaming storage substrate behind all of them. Read the full breakdown to see which fits your stack. See the full comparison→ CommunityBuilt in the open, governed by the ASF. Apache Fluss is developed openly by a global community of contributors. Join the discussion, file an issue, or send a patch. Apache 2.0 Open-source license ASF Apache Software Foundation governance GitHub Source code, issues, and pull requests. Open repository → Slack Real-time chat with users and committers. Join the workspace → Contribute Welcome guide, mailing lists, and how to send your first patch. Get started → Product Documentation Quickstart Roadmap Downloads Blog Community GitHub Slack Welcome Contribute Resources Talks Videos Issues Releases Apache Foundation License Events Donate Sponsors Security Privacy Copyright © 2026 The Apache Software Foundation, Licensed under the Apache License, Version 2.0. Apache, the names of Apache projects, and the feather logo are either registered trademarks or trademarks of the Apache Software Foundation in the United States and/or other countries. All other marks mentioned may be trademarks or registered trademarks of their respective owners.