Apache SeaTunnel (Incubating)

SeaTunnel was formerly named Waterdrop , and renamed SeaTunnel since October 12, 2021.

SeaTunnel is a very easy-to-use ultra-high-performance distributed data integration platform that supports real-time synchronization of massive data. It can synchronize tens of billions of data stably and efficiently every day, and has been used in the production of nearly 100 companies.

Why do we need SeaTunnel

SeaTunnel focuses on data integration and data synchronization, and is mainly designed to solve common problems in the field of data integration:

Various data sources: There are hundreds of commonly-used data sources of which versions are incompatible. With the emergence of new technologies, more data sources are appearing. It is difficult for users to find a tool that can fully and quickly support these data sources.
Complex synchronization scenarios: Data synchronization needs to support various synchronization scenarios such as offline-full synchronization, offline-incremental synchronization, CDC, real-time synchronization, and full database synchronization.
High demand in resource: Existing data integration and data synchronization tools often require vast computing resources or JDBC connection resources to complete real-time synchronization of massive small tables. This has increased the burden on enterprises to a certain extent.
Lack of quality and monitoring: Data integration and synchronization processes often experience loss or duplication of data. The synchronization process lacks monitoring, and it is impossible to intuitively understand the real-situation of the data during the task process.
Complex technology stack: The technology components used by enterprises are different, and users need to develop corresponding synchronization programs for different components to complete data integration.
Difficulty in management and maintenance: Limited to different underlying technology components (Flink/Spark) , offline synchronization and real-time synchronization often have be developed and managed separately, which increases the difficulty of the management and maintainance.

Features of SeaTunnel

Rich and extensible Connector: SeaTunnel provides a Connector API that does not depend on a specific execution engine. Connectors (Source, Transform, Sink) developed based on this API can run on many different engines, such as SeaTunnel Engine, Flink, Spark that are currently supported.
Connector plugin: The plugin design allows users to easily develop their own Connector and integrate it into the SeaTunnel project. Currently, SeaTunnel has supported more than 70 Connectors, and the number is surging. There is the list of connectors we supported and plan to support.
Batch-stream integration: Connectors developed based on SeaTunnel Connector API are perfectly compatible with offline synchronization, real-time synchronization, full- synchronization, incremental synchronization and other scenarios. It greatly reduces the difficulty of managing data integration tasks.
Support distributed snapshot algorithm to ensure data consistency.
Multi-engine support: SeaTunnel uses SeaTunnel Engine for data synchronization by default. At the same time, SeaTunnel also supports the use of Flink or Spark as the execution engine of the Connector to adapt to the existing technical components of the enterprise. In addition, SeaTunnel supports multiple versions of Spark and Flink.
JDBC multiplexing, database log multi-table parsing: SeaTunnel supports multi-table or whole database synchronization, which solves the problem of over-JDBC connections; supports multi-table or whole database log reading and parsing, which solves the need for CDC multi-table synchronization scenarios problems with repeated reading and parsing of logs.
High throughput and low latency: SeaTunnel supports parallel reading and writing, providing stable and reliable data synchronization capabilities with high throughput and low latency.
Perfect real-time monitoring: SeaTunnel supports detailed monitoring information of each step in the data synchronization process, allowing users to easily understand the number of data, data size, QPS and other information read and written by the synchronization task.
Two job development methods are supported: coding and canvas design. The SeaTunnel web project https://github.com/apache/incubator-seatunnel-web provides visual management of jobs, scheduling, running and monitoring capabilities.

SeaTunnel work flowchart

The runtime process of SeaTunnel is shown in the figure above.

The user configures the job information and selects the execution engine to submit the job.

The Source Connector is responsible for parallelizing the data and sending the data to the downstream Transform or directly to the Sink, and the Sink writes the data to the destination. It is worth noting that both Source and Transform and Sink can be easily developed and extended by yourself.

The default engine use by SeaTunnel is SeaTunnel Engine. If you choose to use the Flink or Spark engine, SeaTunnel will package the Connector into a Flink or Spark program and submit it to Flink or Spark to run.

Connectors supported by SeaTunnel

Source Connectors supported check out
Sink Connectors supported check out
Transform supported check out

Here's a list of our connectors with their health status.connector status

Environmental dependency

java runtime environment, java >= 8
If you want to run SeaTunnel in a cluster environment, any of the following Spark cluster environments is usable:

Spark on Yarn
Spark Standalone

If the data volume is small, or the goal is merely for functional verification, you can also start in local mode without a cluster environment, because SeaTunnel supports standalone operation. Note: SeaTunnel 2.0 supports running on Spark and Flink.

Compiling project

Follow this document.

Downloads

Download address for run-directly software package : https://seatunnel.apache.org/download

Quick start

SeaTunnel Engine https://seatunnel.apache.org/docs/start-v2/locally/quick-start-seatunnel-engine/

Spark https://seatunnel.apache.org/docs/start-v2/locally/quick-start-spark

Flink https://seatunnel.apache.org/docs/start-v2/locally/quick-start-flink

Application practice cases

Weibo, Value-added Business Department Data Platform

Weibo business uses an internal customized version of SeaTunnel and its sub-project Guardian for SeaTunnel On Yarn task monitoring for hundreds of real-time streaming computing tasks.

Sina, Big Data Operation Analysis Platform

Sina Data Operation Analysis Platform uses SeaTunnel to perform real-time and offline analysis of data operation and maintenance for Sina News, CDN and other services, and write it into Clickhouse.

Sogou, Sogou Qiqian System

Sogou Qiqian System takes SeaTunnel as an ETL tool to help establish a real-time data warehouse system.

Qutoutiao, Qutoutiao Data Center

Qutoutiao Data Center uses SeaTunnel to support mysql to hive offline ETL tasks, real-time hive to clickhouse backfill technical support, and well covers most offline and real-time tasks needs.

Yixia Technology, Yizhibo Data Platform
Yonghui Superstores Founders' Alliance-Yonghui Yunchuang Technology, Member E-commerce Data Analysis Platform

SeaTunnel provides real-time streaming and offline SQL computing of e-commerce user behavior data for Yonghui Life, a new retail brand of Yonghui Yunchuang Technology.

Shuidichou, Data Platform

Shuidichou adopts SeaTunnel to do real-time streaming and regular offline batch processing on Yarn, processing 3~4T data volume average daily, and later writing the data to Clickhouse.

Tencent Cloud

Collecting various logs from business services into Apache Kafka, some of the data in Apache Kafka is consumed and extracted through Seatunnel, and then store into Clickhouse.

For more use cases, please refer to: https://seatunnel.apache.org/blog

Code of conduct

This project adheres to the Contributor Covenant code of conduct. By participating, you are expected to uphold this code. Please follow the REPORTING GUIDELINES to report unacceptable behavior.