Gravity is used to replicate data between various inputs and outputs (databases, message queues).

It is designed to be a customizable data replication tool that:

• Supports multiple sources and destinations

• Supports Kubernetes-based replication cluster

• Data Bus: Use Change Data Capture (MySQL binlog, MongoDB Oplog) and batch table scan to publish data to message queue like Kafka.
• Unidirectional data replication: Replicates data from one MySQL cluster to another MySQL cluster.
• Bidirectional data replication: Replicates data between two MySQL clusters bidirectionally.
• Synchronization of shards to the merged table: Synchronizes MySQL sharded tables to the merged table. You can specify the corresponding relationship between the source table and the target table.
• Online data mutation: Data can be changed during the replication. For example, rename the column, encrypt/decrypt data columns.

Gravity supports the following data mutations:

• Ignore specific data
• Renaming columns
• Deleting columns

• Architecture
• Quick Start
• Configuration
• Cluster Deployment

Special thanks to @dantin, @LiangShang, and @liwink for the early support to this project, and thanks @CaitinChen for the initial translation of the documentation.