Apache StreamPark
An Apache Software Foundation project maintained by its open-source community- Application development, deployment and operations
- Apache Flink multi-version support
- Community releases and documentation
Discuss catalogs, AI access, processing, lakehouse integrations and support. The proposal specifies edition entitlements, component versions, deployment responsibilities and services.
Commercial features and services depend on the selected release and agreement. Confirm supported components, maintenance term and service hours in the proposal. Additional integration and customization are scoped separately.
Compare 42 capabilities across deployment, development, operations, and governance.
Apache StreamPark is an independent Apache project, not a free Realfuture edition. Community installation references use the documented 2.1.5 examples; capabilities vary by release. This is a guide to product scope, not a release certification matrix. Community installation documentation · Compatibility and evaluation guide
| Category | Capability | 01 / OPEN SOURCEApache StreamPark | 02 / REALFUTUREEnterprise |
|---|---|---|---|
| 01Software | Platform and components | Connects to supported Apache Flink runtimes. Flink and Flink CDC are maintained by their own communities. | AI-driven streaming lakehouse platform, with enhancements to catalogs, Flink, CDC and ecosystem integrations. The delivered release defines enhancements, component versions and maintenance scope. |
| Development and operations | Application development, deployment and operations, including Flink SQL and JAR workflows. | Shared catalogs, AI-assisted SQL/CDC/ETL, queries and validation, JAR/Git publishing and production operations. AI requires configured model access and the agreed edition. Query modes, ETL topology and publishing depend on the selected release. | |
| 02Deployment | Installation and platform database | Self-hosted installation, including Docker Compose. MySQL and PostgreSQL are documented platform database options. | Deployment planning and platform enhancements for the customer environment. Platform HA, database adaptations and installation responsibilities are scoped for the deployment. |
| Runtime compatibility | Supports multiple Flink versions and deployment modes, depending on the StreamPark release. | Management of connected runtimes, with commercial enhancements for selected components. Confirm Flink, CDC, JDK, connector and infrastructure versions together. Enhanced features may require component upgrades. | |
| 03Integration | Connectors and ecosystem | Uses connectors and ecosystem components available for the selected runtime. | Catalog and connector integrations for Paimon, Fluss, Doris and StarRocks, with component-specific development and maintenance workflows. Source and sink directions, data types and schema-change behavior depend on the connector and version. |
| Enterprise systems | Community capabilities and integration options vary by release. | Platform integration and additional engineering services can be included in the proposal. Agree on identity, API, monitoring and other integration requirements before a proof of concept. | |
| 04Support | Maintenance and assistance | Community releases, public issue tracking and community assistance. | Enterprise technical support and maintenance for the agreed software scope. Covered versions, maintenance term, service hours and response targets are defined in the agreement. |
| Migration and delivery | Documentation and community resources; deployment and validation remain with the operator. | Deployment, migration, training and tuning services can be scoped with the team. Validate representative jobs, state compatibility and rollback steps in the target environment. |
Commercial scope follows the operating responsibilities Realfuture takes on.
Data systems, catalogs, workloads, AI access, teams and production criticality.
Private cloud, customer cloud, hybrid topology, and operational ownership.
Response expectations, support lifecycle, migration, and production assistance.
Optional connectors, integration, training, and scoped customization.
Share job count, runtime clusters, deployment boundary, and support needs. We’ll recommend the right tier and define the inputs for a scoped proposal.