SeaTunnel Connectors Capability Overview

SeaTunnel provides a comprehensive set of connectors that enable you to read from various data sources and write to different data sinks. This document provides a detailed capability matrix for all available connectors based on the Connector V2 Features.

Quick Facts

• Total Source Connectors: 79
• Total Sink Connectors: 78
• Total Connectors: 157
• Supported Engines: Spark, Flink, SeaTunnel Zeta
• Supported Data Types: Structured, Unstructured, Multimodal

Feature Definitions

Source Connector Features

Feature	Description
exactly-once	Each piece of data is sent downstream only once, with state snapshots and offsets for reliability
column projection	Read only specified columns from data source efficiently
batch	Supports bounded data processing (job stops after completing all data)
stream	Supports unbounded data processing (continuous streaming)
parallelism	Supports parallel execution with multiple tasks reading different splits
multimodal	Supports structured and unstructured data (text, video, images, binary files)
support user-defined split	Users can configure custom split rules
support multiple table read	Read multiple tables in one SeaTunnel job

Sink Connector Features

Feature	Description
exactly-once	Each piece of data is written to target only once via key deduplication or XA transactions
cdc	Supports change data capture with INSERT/UPDATE/DELETE operations based on primary key
support multiple table write	Write to multiple tables in one SeaTunnel job with dynamic table identifiers
multimodal	Supports structured and unstructured data (text, video, images, binary files)

Source Connectors Capability Matrix

Database & CDC Connectors

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
Jdbc	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
MySQL	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
PostgreSQL	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Oracle	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
SQLServer	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
DB2	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Kingbase	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Hive	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
HiveJdbc	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Clickhouse	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Doris	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
StarRocks	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Phoenix	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Greenplum	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Redshift	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Vertica	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
MySQL-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
PostgreSQL-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
Oracle-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
SQLServer-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
TiDB-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
MongoDB-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅
Opengauss-CDC	❌	✅	✅	✅	❌	❌	✅	✅	✅	✅	✅

NoSQL Databases

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
MongoDB	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Cassandra	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	❌
Hbase	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Redis	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Neo4j	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌

Data Lake & Warehouse

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
Iceberg	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Hudi	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Paimon	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Databend	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Maxcompute	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
OceanBase	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅

Message Queues

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
Kafka	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	✅
Pulsar	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	✅
Rabbitmq	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	✅
RocketMQ	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	✅
AmazonSqs	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	✅

File Systems

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
LocalFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
HdfsFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
S3File	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
OssFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
OssJindoFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
ObsFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
CosFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
FtpFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅
SftpFile	✅	✅	✅	✅	✅	✅	✅	✅	✅	✅

Time Series & Search Engines

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
InfluxDB	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
IoTDB	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
IoTDBv2	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
TDengine	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Elasticsearch	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	✅
Easysearch	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Typesense	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Prometheus	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌

Vector Databases

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
Milvus	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Qdrant	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌

APIs & Cloud Services

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
Http	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Socket	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Github	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Gitlab	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Jira	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Notion	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
GoogleSheets	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
GraphQL	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
AmazonDynamoDB	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Klaviyo	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Lemlist	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
MyHours	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
OneSignal	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Persistiq	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Web3j	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌

Special & Test Connectors

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	column projection	batch	stream	parallelism	multimodal	user-defined split	multiple table
FakeSource	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	❌
Cloudberry	✅	✅	✅	✅	✅	✅	✅	✅	❌	❌	❌
Kudu	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
OpenMldb	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌
Tablestore	✅	✅	✅	✅	❌	✅	✅	✅	❌	❌	❌

Sink Connectors Capability Matrix

Database Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Jdbc	✅	✅	✅	✅	❌	✅	❌
MySQL	✅	✅	✅	✅	❌	❌	✅
PostgreSQL	✅	✅	✅	✅	❌	❌	✅
Oracle	✅	✅	✅	✅	❌	❌	✅
SQLServer	✅	✅	✅	✅	❌	❌	✅
DB2	✅	✅	✅	✅	❌	❌	✅
Kingbase	✅	✅	✅	✅	❌	❌	✅
Hive	✅	✅	✅	✅	❌	❌	✅
Phoenix	✅	✅	✅	✅	❌	❌	✅
Greenplum	✅	✅	✅	✅	❌	❌	✅

NoSQL & Graph Databases

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
MongoDB	✅	✅	✅	✅	❌	❌	❌
Cassandra	✅	✅	✅	✅	❌	❌	❌
Hbase	✅	✅	✅	✅	❌	❌	❌
Redis	✅	✅	✅	✅	❌	❌	❌
Neo4j	✅	✅	✅	✅	❌	❌	❌
Aerospike	✅	✅	✅	✅	❌	❌	❌
AmazonDynamoDB	✅	✅	✅	✅	❌	❌	❌
GoogleFirestore	✅	✅	✅	✅	❌	❌	❌
HugeGraph	✅	✅	✅	✅	❌	❌	❌

Data Warehouse & Analytical Databases

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Clickhouse	✅	✅	✅	✅	❌	❌	❌
Doris	✅	✅	✅	✅	❌	❌	❌
StarRocks	✅	✅	✅	✅	❌	❌	❌
Redshift	✅	✅	✅	✅	❌	❌	❌
Snowflake	✅	✅	✅	✅	❌	❌	❌
Databend	✅	✅	✅	✅	❌	❌	❌
Vertica	✅	✅	✅	✅	❌	❌	❌
Druid	✅	✅	✅	✅	❌	❌	❌

Message Queue Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Kafka	✅	✅	✅	✅	❌	❌	❌
Pulsar	✅	✅	✅	✅	❌	❌	❌
Rabbitmq	✅	✅	✅	✅	❌	❌	❌
RocketMQ	✅	✅	✅	✅	❌	❌	❌
AmazonSqs	✅	✅	✅	✅	❌	❌	❌
Activemq	✅	✅	✅	✅	❌	❌	❌

File System Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
LocalFile	✅	✅	✅	✅	❌	❌	❌
HdfsFile	✅	✅	✅	✅	❌	❌	❌
S3File	✅	✅	✅	✅	❌	❌	❌
OssFile	✅	✅	✅	✅	❌	❌	❌
OssJindoFile	✅	✅	✅	✅	❌	❌	❌
ObsFile	✅	✅	✅	✅	❌	❌	❌
CosFile	✅	✅	✅	✅	❌	❌	❌
FtpFile	✅	✅	✅	✅	❌	❌	❌
SftpFile	✅	✅	✅	✅	❌	❌	❌
ClickhouseFile	✅	✅	✅	✅	❌	❌	❌

Data Lake Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Iceberg	✅	✅	✅	✅	❌	❌	❌
Hudi	✅	✅	✅	✅	❌	❌	❌
Paimon	✅	✅	✅	✅	❌	❌	❌

Search & Time Series Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Elasticsearch	✅	✅	✅	✅	❌	❌	❌
Easysearch	✅	✅	✅	✅	❌	❌	❌
Typesense	✅	✅	✅	✅	❌	❌	❌
InfluxDB	✅	✅	✅	✅	❌	❌	❌
IoTDB	✅	✅	✅	✅	❌	❌	❌
IoTDBv2	✅	✅	✅	✅	❌	❌	❌
TDengine	✅	✅	✅	✅	❌	❌	❌
Prometheus	✅	✅	✅	✅	❌	❌	❌

Vector Database Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Milvus	✅	✅	✅	✅	❌	❌	❌
Qdrant	✅	✅	✅	✅	❌	❌	❌

API & Cloud Service Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Http	✅	✅	✅	✅	❌	❌	❌
GraphQL	✅	✅	✅	✅	❌	❌	❌
Socket	✅	✅	✅	✅	❌	❌	❌
Datahub	✅	✅	✅	✅	❌	❌	❌
Maxcompute	✅	✅	✅	✅	❌	❌	❌

Specialized Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
Console	✅	✅	✅	❌	❌	❌	❌
Assert	✅	✅	✅	❌	❌	❌	❌
Email	✅	✅	✅	❌	❌	❌	❌
Slack	✅	✅	✅	❌	❌	❌	❌
DingTalk	✅	✅	✅	❌	❌	❌	❌
Feishu	✅	✅	✅	❌	❌	❌	❌
Enterprise-WeChat	✅	✅	✅	❌	❌	❌	❌
Sentry	✅	✅	✅	❌	❌	❌	❌
SensorsData	✅	✅	✅	❌	❌	❌	❌

Other Sinks

Connector	Spark	Flink	SeaTunnel Zeta	exactly-once	cdc	multiple table	multimodal
S3-Redshift	✅	✅	✅	✅	❌	❌	❌
SelectDB-Cloud	✅	✅	✅	✅	❌	❌	❌
Tablestore	✅	✅	✅	✅	❌	❌	❌
Kudu	✅	✅	✅	✅	❌	❌	❌
Cloudberry	✅	✅	✅	✅	❌	❌	❌
Fluss	✅	✅	✅	✅	❌	❌	❌
OceanBase	✅	✅	✅	✅	❌	❌	❌

Feature Support Summary

Source Connectors Feature Distribution

Feature	Count	Percentage	Examples
Engine Support
All Three Engines	~65	82%	Most database, file, and message queue connectors
Spark + Flink Only	~8	10%	Some specialized API connectors
Flink + Zeta Only	~4	5%	CDC connectors
Processing Mode
Batch Support	~60	76%	Most database and file connectors
Stream Support	~70	89%	Most connectors, especially CDC and messaging
Reliability
Exactly-Once	~45	57%	File connectors, JDBC, Kafka
Performance
Parallelism	~55	70%	Most database and file connectors
Column Projection	~25	32%	JDBC, File, some specialized connectors
Advanced Features
User-Defined Split	~15	19%	CDC, some file connectors
Multiple Table Read	~25	32%	JDBC and some database connectors
Multimodal Support	~10	13%	File and some specialized connectors

Sink Connectors Feature Distribution

Feature	Count	Percentage	Examples
Engine Support
All Three Engines	~70	90%	Most database and file connectors
Spark + Flink Only	~6	8%	Some specialized connectors
Flink + Zeta Only	~2	3%	Specialized cases
Reliability
Exactly-Once	~40	51%	JDBC, Kafka, File connectors
Data Capabilities
CDC Support	~5	6%	Limited to specialized database sinks
Multiple Table Write	~15	19%	JDBC and some database sinks
Multimodal Support	~10	13%	File and specialized connectors

Connector Selection Guide

Use Case-Based Recommendations

For High Throughput Batch Processing

• Recommended Sources: Jdbc, File connectors (LocalFile, HdfsFile, S3File)
• Recommended Sinks: Jdbc, File connectors, Data Lake formats (Iceberg, Hudi)
• Key Features: Batch support, parallelism, column projection

For Real-time Stream Processing

• Recommended Sources: CDC connectors, Kafka, File connectors (stream mode)
• Recommended Sinks: Kafka, Jdbc (with transactions), Real-time databases
• Key Features: Stream support, exactly-once, low latency

For Exactly-Once Guarantees

• Recommended Sources: File connectors, JDBC, Kafka
• Recommended Sinks: JDBC (XA transactions), Kafka (2PC), File connectors
• Key Features: Exactly-once, transaction support, state management

For Multi-Table Operations

• Recommended Sources: JDBC connectors with multi-table support
• Recommended Sinks: JDBC with dynamic table identifiers
• Key Features: Multiple table read/write, placeholder support

For Cloud Integration

• Recommended Sources: Native cloud connectors, File connectors with cloud storage
• Recommended Sinks: Cloud-specific connectors, File connectors
• Examples: S3File, OSSFile, Snowflake, Redshift, MaxCompute

For Advanced Analytics

• Recommended Sources: Data lake formats, Analytical databases
• Recommended Sinks: Data lake formats (Iceberg, Hudi, Paimon), OLAP databases
• Examples: Clickhouse, Doris, StarRocks, Druid

Engine Compatibility Notes

SeaTunnel Zeta (Recommended)

• Advantages: Best performance, most features, unified API
• Connector Coverage: ~82% source, ~90% sink
• Use Cases: Production deployments, performance-critical workloads

Apache Flink

• Advantages: Stream processing excellence, fault tolerance
• Connector Coverage: ~95% source, ~98% sink
• Use Cases: Complex streaming, stateful processing

Apache Spark

• Advantages: Batch processing, ecosystem integration
• Connector Coverage: ~90% source, ~98% sink
• Use Cases: Large-scale batch processing, ETL workflows

Data Format Support

Data Format	Source Support	Sink Support	Primary Connectors
JSON	✅ Most	✅ Most	Universal default format
CSV	✅ File	✅ File	LocalFile, HdfsFile, S3File
Avro	✅ Kafka/File	✅ Kafka/File	Kafka, File connectors
Parquet	✅ File/Hive	✅ File/Hive	LocalFile, HdfsFile, Hive
ORC	✅ File/Hive	✅ File/Hive	LocalFile, HdfsFile, Hive
Text	✅ File/Kafka	✅ File/Kafka	File connectors, Kafka
XML	✅ File	✅ File	File connectors
Protobuf	✅ Kafka	✅ Kafka	Kafka
Canal-JSON	✅ Kafka	✅ Kafka	Kafka
Debezium-JSON	✅ Kafka	✅ Kafka	Kafka
Maxwell-JSON	✅ Kafka	✅ Kafka	Kafka
OGG-JSON	✅ Kafka	✅ Kafka	Kafka

Getting Started

Quick Setup

1. Choose Engine: Select SeaTunnel Zeta for best performance
2. Select Connectors: Use the matrices above to choose appropriate source/sink
3. Install Plugins: Download required connector JAR files
4. Configure Job: Create configuration based on feature requirements
5. Test & Deploy: Validate configuration and run production jobs

Best Practices

1. Feature Matching: Choose connectors that support your required features
2. Engine Selection: Use SeaTunnel Zeta when possible for maximum compatibility
3. Performance: Enable parallelism and batch processing where supported
4. Reliability: Prioritize exactly-once support for critical workloads
5. Monitoring: Monitor connector performance and adjust configurations

Contributing

Want to add new connectors or improve existing ones? Check our:

• Contributor Guide
• Connector Development Guidelines
• Community Forums

---

This matrix represents the current state of SeaTunnel connectors based on official documentation. For the most up-to-date information, refer to individual connector documentation pages. Feature availability may vary between versions.