This topic describes the selection and documentation entry points for migrating data to OceanBase Database (or between OceanBase tenants, or out to other databases) by using ecosystem tools such as OMS, obloader, and Flink. For heterogeneous scenarios, pay attention to schema and type mapping, parallelism and batching strategies, and data verification after cutover. When integrating with data warehouses and files, also consider format conversion and metadata alignment.
Migrate data to OceanBase by using OMS
OMS (OceanBase Migration Service) provides integrated orchestration from schema migration, full migration to incremental migration. It supports various combinations of relational source databases and OceanBase tenant types, and has task monitoring and switchover capabilities. It is suitable for enterprise-level database replacement and long-term synchronization projects.
Description |
Link |
|---|---|
| OMS Documentation | OMS Documentation |
Migrate from other databases to OceanBase
Between OceanBase clusters / tenants (OMS)
Scenario |
Documentation |
|---|---|
| Migration between tenants of the same type | Use OMS to migrate data from OceanBase Database to a tenant of the same type in OceanBase Database |
| Disaster Recovery and Active-Active Architecture for Tenants of the Same Type | Create an active-active disaster recovery project for tenants of the same type in OceanBase Database by using OMS |
Big data sources (OMS Community Edition)
Source |
Documentation |
|---|---|
| Hive | Use OMS Community Edition to migrate Hive data |
| HBase | Use OMS Community Edition to migrate HBase data |
Migrate data to OceanBase by using obloader
obloader is suitable for file import: it can batch write files such as CSV, Parquet, ORC, and SQL from local disks, object storage, or HDFS to OceanBase. When used with direct load, it can improve throughput. It is typically used in scenarios where offline file exports are ready, and an import tool is needed to read the files, parse the formats, and write the data to OceanBase. If data files are network-isolated from OBServer nodes, obloader can also perform data import.
Scenario |
Documentation |
|---|---|
| Obloader Product Documentation | obloader Documentation |
| Migrate from Hive | Use obloader to migrate Hive data |
| ORC / Parquet / CSV files | Use obloader to migrate files in ORC, Parquet, or CSV formats |
| Between OceanBase tenants (data migration tools) | See the following table. |
Migrate data between OceanBase tenants by using OceanBase data migration tools
Scenario |
Documentation |
|---|---|
| MySQL-compatible tenant → MySQL-compatible tenant | Use OceanBase Data Migration Tools to migrate data between tenants in MySQL-compatible mode of an OceanBase cluster |
| MySQL-compatible tenant → Oracle-compatible tenant | Use OceanBase data migration tools to migrate data from a MySQL-compatible mode tenant to an Oracle-compatible mode tenant in an OceanBase cluster |
| Oracle-compatible tenant → Oracle-compatible tenant | Use OceanBase data migration tools to migrate data between tenants in Oracle compatible mode of an OceanBase cluster |
| Oracle-compatible tenant → MySQL-compatible tenant | Use the OceanBase data migration tool to migrate data from an Oracle-compatible mode tenant to a MySQL-compatible mode tenant in an OceanBase cluster |
| Direct load mode (RPC, network isolation, etc.) | Import data by using obloader |
Migrate data to OceanBase by using ecosystem tools
Migrate from MySQL to OceanBase
Migrate from Oracle to OceanBase
Migrate from DB2 LUW to OceanBase
tools |
Documentation |
|---|---|
| DBCAT (Table Schema) | Use DBCAT to migrate table structures from a DB2 LUW database to OceanBase Database |
Spark and Flink related
Scenario |
Documentation |
|---|---|
| Read and Write OceanBase Database from Spark | OceanBase Spark Connector |
| StarRocks (inconsistent with upstream RDS, processed by Flink, etc.) | Use Flink-OMT to migrate data from StarRocks |