Use WarehousePG Backup and Restore (whpg-backup) to create and restore logical backups of your WarehousePG tables in parallel. By default, backups are written to local disk on the coordinator and segment hosts, though you can consolidate them in a shared directory or write them directly to S3-compatible storage. whpg-backup writes the metadata and DDL for a backup to the coordinator host, while each segment writes its own table data to CSV files, so backup and restore performance scales with your cluster.
Key capabilities
- Parallel backup and restore across the coordinator and all segment hosts
- Full database backups, plus incremental backups that include every heap table but back up append-optimized tables only if they changed
- Filtering backups and restores by schema, table, or leaf partition
- Email notifications when a backup or restore completes
- Backing up directly to Amazon S3 or S3-compatible storage
Release notes
Release notes provide information on what's new in each release of WarehousePG Backup and Restore.
Overview
Learn about the architecture, compatibility, and known issues for WarehousePG Backup and Restore.
Installing
Learn how to install the gpbackup and gprestore utilities on your WarehousePG cluster.
Backing up and restoring
Back up and restore WarehousePG tables with gpbackup and gprestore, including full backups, filtering, email notifications, and incremental backups.
Creating incremental backups
Create and restore incremental backups of append-optimized tables with gpbackup and gprestore.
Using the S3 storage plugin
Configure gpbackup and gprestore to back up WarehousePG tables directly to Amazon S3 or an S3-compatible storage server.
Reference
The complete reference to gpbackup and gprestore commands.