DEV.co
Open-Source Databases · risingwavelabs

risingwave

RisingWave is an open-source event streaming platform written in Rust that ingests data from databases, event streams, and webhooks, processes it continuously with sub-100ms freshness, and serves results via SQL. It aims to replace the traditional stack of separate tools (Debezium, Kafka, Flink, database) with a single unified system.

Source: GitHub — github.com/risingwavelabs/risingwave
9.2k
GitHub stars
789
Forks
Rust
Primary language
Apache-2.0
License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

FieldValue
Repositoryrisingwavelabs/risingwave
Ownerrisingwavelabs
Primary languageRust
LicenseApache-2.0 — OSI-approved
Stars9.2k
Forks789
Open issues1.6k
Latest releasev3.0.1 (2026-06-30)
Last updated2026-07-08
Sourcehttps://github.com/risingwavelabs/risingwave

What risingwave is

RisingWave performs incremental computation over streaming data using materialized views, stores state in object storage (S3-compatible) for cost efficiency, and serves query results from an in-memory row store at 10–20ms p99 latency. It supports Apache Iceberg for durable, open-format storage, integrates via PostgreSQL wire protocol, and provides connectors for Kafka, Pulsar, Kinesis, webhooks, and CDC from PostgreSQL/MySQL.

Quickstart

Get the risingwave source

Clone the repository and explore it locally.

terminalbash
git clone https://github.com/risingwavelabs/risingwave.gitcd risingwave# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Real-time feature stores

Compute batch and streaming features over the same pipeline and serve them at low latency directly from RisingWave, eliminating the need for separate feature serving infrastructure.

Live dashboards and monitoring

Maintain continuously updated materialized views that reflect upstream data changes within 100ms, enabling real-time dashboards and alert systems without scheduled refreshes or polling.

Streaming lakehouses with open-format persistence

Ingest event streams into Apache Iceberg tables with automated compaction and snapshot management, keeping data queryable by any Iceberg-compatible engine (Spark, Trino, DuckDB) while serving hot queries from the row store.

Implementation considerations

  • RisingWave requires object storage (S3 or compatible) for state durability; ensure your infrastructure supports this and plan for egress costs.
  • Query latency depends on whether hot data fits in local disk cache; plan disk/SSD sizing for your workload's working set.
  • Webhook ingestion and CDC connectors require stable, low-latency connections to upstream sources; network reliability directly affects end-to-end freshness.
  • Apache Iceberg integration is built-in but requires an accessible REST catalog; plan for catalog availability and performance.
  • Telemetry is enabled by default (anonymized); review documentation and disable if required by your compliance policies.

When to avoid it — and what to weigh

  • Batch-only workloads — RisingWave is optimized for continuous, incremental processing. If your use case requires only scheduled batch ingestion and transformation, simpler, lighter-weight tools may be more cost-effective.
  • Strict air-gapped or on-premise-only deployments — RisingWave's architecture relies on object storage (S3 or equivalent) for state. Air-gapped environments require careful planning and may require architectural workarounds.
  • Sub-microsecond latency requirements — While sub-100ms freshness is excellent for most use cases, applications requiring microsecond-scale latency should evaluate specialized streaming systems.
  • High-cardinality, unbounded state scenarios — Workloads with very large or unbounded state (e.g., unbounded windowed aggregations across millions of keys) may incur significant object storage I/O and cost.

License & commercial use

Apache License 2.0 (Apache-2.0). This is a permissive OSI-approved license allowing modification, distribution, and commercial use, subject to attribution and inclusion of the license text in distributions.

Apache License 2.0 permits commercial use without explicit permission, provided you retain attribution and include the license in derivative works or distributions. No vendor lock-in or commercial license required. However, review license compliance requirements for your specific deployment and derivative works.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

SignalAssessment
MaintenanceActive
DocumentationAdequate
License clarityClear
Deployment complexityModerate
DEV.co fitGood
Assessment confidenceHigh
Security considerations

No exploit details disclosed. Key considerations: (1) PostgreSQL wire protocol exposure requires network isolation and authentication controls; (2) object storage credentials must be managed securely; (3) webhook ingestion may accept untrusted inputs—validate and rate-limit upstream sources; (4) telemetry is enabled by default—review privacy requirements; (5) no explicit mention of encryption at rest or in transit in README—requires review of operational documentation; (6) standard OSS security: review commit history and dependencies for known vulnerabilities.

Alternatives to consider

Apache Flink

Mature, widely-adopted streaming framework with comprehensive operator ecosystem. Steeper operational overhead (cluster management, state backend configuration) but battle-tested at scale.

Kafka Streams / Spring Cloud Stream

Lightweight, library-based streaming for JVM applications. Suitable for event-driven microservices; less centralized than RisingWave and requires embedding in application code.

Databend / ClickHouse

OLAP databases with streaming ingestion support. Optimized for analytics at scale rather than incremental serving; different trade-off between freshness and analytical depth.

Software development agency

Build on risingwave with DEV.co software developers

RisingWave may be a strong fit if you need low-latency, continuously updated data serving, unified ingestion from multiple sources, or a simpler alternative to Kafka + Flink + database stacks. Start with the quick-start guide and benchmark your workload.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

Related on DEV.co

Explore the category and the services that help you build with it.

risingwave FAQ

Does RisingWave replace Kafka?
No. RisingWave consumes from Kafka (and other event streams) as a source. It replaces the downstream processing and serving layers (Flink + database), not Kafka itself.
How does RisingWave achieve sub-100ms freshness?
It performs incremental computation: when upstream data changes, only affected results are recomputed rather than full recalculation. State is maintained in object storage and served from an in-memory row store.
Is RisingWave suitable for small teams?
Managed RisingWave Cloud simplifies operations for small teams. Self-hosted requires object storage, disk caching setup, and monitoring; use Kubernetes for easier scaling or Docker Compose for single-node testing.
Can I query historical data from Iceberg while serving fresh results?
Yes. The row store serves low-latency queries on fresh data; Iceberg stores durable, long-term data. Queries can span both via standard SQL, with RisingWave routing to appropriate layers.

Custom software development services

From first prototype to production, DEV.co delivers software development services around tools like risingwave. Our software development agency staffs experienced software developers and web developers for custom software development, web development, integrations, and ongoing support across open-source databases and beyond.

Evaluate RisingWave for your streaming data pipeline

RisingWave may be a strong fit if you need low-latency, continuously updated data serving, unified ingestion from multiple sources, or a simpler alternative to Kafka + Flink + database stacks. Start with the quick-start guide and benchmark your workload.