TiDB to Amazon S3 Data Lake Integration

How Datrise loads TiDB into Amazon S3 Data Lake

Datrise syncs TiDB's tables, rows, schemas, indexes, regions, and TiCDC change events into Amazon S3 Data Lake as columnar Parquet objects partitioned per source entity. Flexible or custom fields land in nested struct/map fields in Parquet, and timestamps such as created, updated, and status changes are typed as ISO-8601 timestamp columns.

Sync is incremental: Datrise uses writes new date partitions and compacts small files on a schedule, so re-runs update only what changed. Hive-style path partitioning (entity/date) for engine-agnostic reads. A lake has no schema enforcement, so Datrise writes a schema manifest alongside the data to keep downstream engines consistent.

Ideal for an open, engine-neutral storage layer for Spark, Athena, Trino, or DuckDB.

Endpoints

TiDB: MySQL-compatible HTAP database with row and column stores.

Amazon S3 Data Lake: Object storage landing zone for parquet and snapshots.

How TiDB entities map to Amazon S3 Data Lake

TiDB entity	Amazon S3 Data Lake object	Notes
tables	`tidb_tables`	id PK · custom fields → nested struct/map fields in Parquet
rows	`tidb_rows`	id PK · linked to tidb_tables
schemas	`tidb_schemas`	id PK · linked to tidb_tables
indexes	`tidb_indexes`	id PK · linked to tidb_tables

FAQ

How does Datrise handle TiDB's custom fields in Amazon S3 Data Lake?

Flexible values are stored as nested struct/map fields in Parquet, so new fields don't require a migration; strongly-typed fields — dates, numbers, and references — are promoted to native Amazon S3 Data Lake types.

How does the TiDB to Amazon S3 Data Lake sync stay up to date?

It runs incrementally — Datrise uses writes new date partitions and compacts small files on a schedule.

Related pipelines

Early access

Connect TiDB to Amazon S3 Data Lake the easy way

Skip brittle scripts and manual exports. Join the waitlist to get a guided setup, AI-assisted mapping, and reliable incremental sync for this integration.