Partner Integration Solutions

Beyond the built-in loading methods, you can load data into VeloDB with solutions from our ecosystem partners. VeloDB is fully compatible with Apache Doris, so every product that ships a connector for Apache Doris works with VeloDB without modification.

This page lists partner solutions that provide a dedicated, officially documented connector for Apache Doris or VeloDB. Tools that can reach the database only through a generic JDBC or MySQL driver are not covered here. Partners are grouped by scenario and listed alphabetically within each group.

Before you start

Most partner connectors write data through Stream Load over HTTP and read metadata over the MySQL protocol. Kafka-compatible streaming platforms are instead consumed by VeloDB itself through Routine Load. Before configuring a partner platform, make sure it can reach the required endpoints of your VeloDB cluster over the network. See Connecting by MySQL Protocol for connection details and credentials.

Solution overview

Partner	Product	Typical scenario	Integration path
Confluent	Confluent Cloud / Platform	Managed Kafka event streaming	Doris Kafka Connector on Kafka Connect, or Routine Load
LaserData	LaserData (Apache Iggy)	Ultra-low-latency streaming	Apache Doris sink connector via Stream Load
Redpanda	Redpanda	High-performance Kafka-compatible streaming	Routine Load over the Kafka API
StreamNative	StreamNative Cloud (Apache Pulsar)	Multi-protocol messaging	Routine Load or Doris Kafka Connector over the Kafka protocol
NineData	NineData	Database DevOps, migration, and replication	Doris source and target for migration and sync
Supermetal	Supermetal	High-performance CDC replication	Doris target, direct or via Apache Iceberg
EMQX	EMQX Enterprise	IoT and MQTT telemetry	Built-in Apache Doris sink in the rule engine
RisingWave	RisingWave	Streaming SQL and incremental materialized views	Native Doris sink, append-only and upsert
CocoIndex	CocoIndex	Real-time indexing for AI workloads	Doris target connector
WhaleOps	WhaleStudio / WhaleTunnel	Enterprise integration and orchestration	Doris connector with batch and CDC sync

Streaming platforms

Confluent

Confluent delivers a fully managed, cloud-native data streaming platform built around Apache Kafka, founded by the original creators of Kafka. Events flowing through Confluent topics can land in VeloDB in two ways: run the Doris Kafka Connector on a Kafka Connect cluster, or let VeloDB pull from topics directly with Routine Load. The connector also applies Debezium CDC records, so inserts, updates, deletes, and schema changes from operational databases arrive correctly.

• Setup guide: Confluent Cloud
• Related reading: Introducing the VeloDB Kafka Connector

LaserData

LaserData is a hyper-efficient streaming platform built on Apache Iggy, written in Rust and engineered around io_uring with a thread-per-core model. Its connector catalog includes an Apache Doris sink that loads messages into Doris tables through Stream Load, so events published to LaserData streams become queryable in VeloDB within milliseconds.

Typical scenario: trading, observability, and other workloads where every millisecond of end-to-end latency counts.

• Partner documentation: LaserData and VeloDB

Redpanda

Redpanda is a high-performance streaming platform that exposes the full Apache Kafka API from a single C++ binary, with no JVM and no ZooKeeper. Because the Kafka API is fully compatible, VeloDB's Routine Load subscribes to Redpanda topics directly, parses JSON, CSV, Avro, or Protobuf, and writes to columnar tables. No Connect cluster or Flink job is required.

Typical scenario: high-throughput event pipelines where predictable tail latency matters, such as trading, telemetry, and clickstream analytics.

• Partner documentation: Redpanda docs

StreamNative

StreamNative, founded by the original creators of Apache Pulsar, provides managed Pulsar clusters with multi-tenancy, geo-replication, and tiered storage. StreamNative brokers speak the Kafka protocol alongside Pulsar, MQTT, and AMQP, so VeloDB consumes StreamNative topics through Routine Load or the Doris Kafka Connector over the Kafka endpoint.

Typical scenario: unifying messaging workloads on one platform while serving real-time analytics from VeloDB.

• Partner documentation: StreamNative Kafka service

CDC and database replication

NineData

NineData is a multi-cloud database DevOps platform that covers SQL development, change management, data migration, replication, and observability across heterogeneous database engines. It supports Doris for both development workflows and data movement, so teams can migrate and replicate data into VeloDB and manage day-to-day SQL work from one console.

Typical scenario: teams operating many database engines that want migration, replication, and governance for VeloDB in one place.

• Partner documentation: Database DevOps for Doris

Supermetal

Supermetal is a high-performance CDC engine for operational databases such as PostgreSQL, MySQL, Oracle, SQL Server, and MongoDB. It ships as a single Rust binary built on Apache Arrow and offers a dedicated Doris target: write changes directly into VeloDB, or land them in Apache Iceberg first and query the lakehouse from VeloDB. No Kafka cluster or Flink job sits in between.

Typical scenario: replicating transactional databases into VeloDB in seconds with minimal infrastructure.

• Partner documentation: Doris target in Supermetal docs

Streaming SQL and IoT

EMQX

EMQX Enterprise is a high-performance MQTT broker built for connected devices and industrial IoT, handling massive concurrent connections with MQTT 5.0 and QoS guarantees. Starting from EMQX Enterprise 5.10, the rule engine ships a built-in Apache Doris sink, so device telemetry can be filtered and transformed with rule SQL and written into VeloDB tables in batches.

Typical scenario: automotive, manufacturing, energy, and smart city telemetry landing directly in VeloDB for real-time dashboards and alerts.

• Partner documentation: Apache Doris sink in EMQX docs

RisingWave

RisingWave is a distributed streaming database that lets you write standard SQL, including joins and aggregations, over streams from Kafka, Pulsar, Kinesis, and more. Its native Doris sink delivers tables and incrementally maintained materialized views into VeloDB, supporting both append-only and upsert modes. Upsert mode requires a Unique Key table on the VeloDB side.

Typical scenario: streaming ETL and pre-aggregation in RisingWave, with serving and ad-hoc analytics in VeloDB.

• Partner documentation: Apache Doris sink in RisingWave docs

AI and data indexing

CocoIndex

CocoIndex is a real-time data indexing framework that watches your sources, transforms records as they change, and writes them into a serving layer, all defined as code. Its Doris connector keeps an always-fresh index in VeloDB, handling incremental updates, dependency tracking, and re-indexing without batch reruns.

Typical scenario: keeping SQL, search, and AI retrieval workloads continuously in sync with changing source data.

• Partner documentation: Doris connector in CocoIndex docs

Orchestration and enterprise integration

WhaleOps

WhaleOps is the team behind Apache DolphinScheduler and Apache SeaTunnel. WhaleStudio combines enterprise-grade workflow orchestration with WhaleTunnel, an integration engine whose connector catalog includes Apache Doris with batch loading, CDC synchronization, and exactly-once delivery. Batch and streaming pipelines run on schedule and land clean, query-ready data in VeloDB.

Typical scenario: enterprise data integration with unified scheduling, monitoring, and governance requirements.

• Partner documentation: WhaleTunnel connectors