RisingWave: Streaming Platform for Agents, Apps & Analytics
What is RisingWave?
Why do teams choose RisingWave for real-time streaming?
Built in Rust, RisingWave combines sub-100 ms real-time data processing with built-in Apache Iceberg integration. It continuously ingests data from databases, message queues, and IoT devices, transforms and enriches it in motion, and materializes results for instant serving. From monitoring and alerting to telemetry, data enrichment, and continuous Iceberg ingestion, RisingWave unifies streaming and lakehouse worlds into a single, consistent platform.
FAQ
What do developers ask about RisingWave?
RisingWave is a distributed SQL streaming platform that ingests, transforms, and serves real-time data using PostgreSQL-compatible SQL. Unlike traditional stream processors, RisingWave has built-in storage and can serve low-latency queries directly — eliminating the need for separate databases, caches, or message queues in your real-time data stack.
How does RisingWave compare to Apache Flink?
RisingWave replaces Flink's Java-based programming model with standard SQL, reducing development time from weeks to hours. It includes built-in state storage (no external RocksDB management), instant failure recovery (vs. minutes-to-hours with Flink), and decoupled compute-storage architecture for independent scaling. Both support exactly-once semantics.
How does RisingWave compare to Apache Kafka?
Kafka is a message broker for data transport. RisingWave is a streaming database that processes and serves data. They are complementary: RisingWave ingests from Kafka, performs complex transformations (joins, aggregations, windowing) via SQL, and serves results at low latency — replacing the need for Kafka Streams, ksqlDB, and separate serving databases.
What are the typical use cases for RisingWave?
RisingWave powers real-time analytics dashboards, fraud detection systems, monitoring and alerting pipelines, IoT telemetry processing, AI agent infrastructure, trading systems, and streaming ETL into data lakes. RisingWave Ultra provides in-memory stream processing for workloads requiring sub-100ms latency. Any workload that requires continuous SQL processing over event streams is a fit for RisingWave.
Is RisingWave compatible with PostgreSQL?
Yes. RisingWave is wire-compatible with PostgreSQL, meaning you can connect using any PostgreSQL client, driver, or tool (psql, JDBC, Python psycopg2, etc.). You write standard SQL to define sources, materialized views, and sinks. There is no new language or API to learn.
Does RisingWave support exactly-once processing?
Yes. RisingWave guarantees exactly-once semantics with consistent snapshots and barrier-based checkpointing. Materialized views are always consistent — even across multi-way joins and complex streaming pipelines. This eliminates the data correctness issues common in eventually-consistent stream processors.
How does RisingWave integrate with Apache Iceberg?
RisingWave Open Lake provides managed Apache Iceberg table maintenance, including streaming ingestion, automatic compaction, and schema evolution. You can continuously sink processed data into Iceberg tables with a single SQL statement, making RisingWave the easiest path from real-time streams to a queryable lakehouse.
Is RisingWave open source?
Yes. RisingWave's core engine is open source under the Apache 2.0 license, available on GitHub with 8,800+ stars. You can self-host it for free. RisingWave Cloud offers a fully managed service with additional enterprise features, SOC 2, GDPR, and HIPAA compliance.