Features

The main feature of the Bacula Enterprise Amazon RDS Plugin is its capability to provide backup and restore for Amazon RDS instances or clusters.

The primary backup layer manages the snapshots of the database instances or clusters, executing them from Bacula Enterprise according to a user-defined schedule and maintaining the desired number of snapshots. On top of this, the plugin can export data to an S3 bucket and to backup the contents of the export operation, resulting in backup jobs that include one Apache Parquet archive for each table, along with several metadata files pertaining to the database and the snapshot. Finally, the plugin is able to manage the complete livecycle of the snapshots created and the exported data. All elements can be retained or cleaned up in a flexible, governed by the configuration settings of Bacula Enterprise.

Apache Parquet is an open source, column-oriented data file format designed for efficient data storage and retrieval. It provides high performance compression and encoding schemes to handle complex data in bulk. For further details, refer to: - https://parquet.apache.org/

Apache Parquet is an independent database format, and the Bacula Enterprise Amazon RDS Plugin offers the ability to restore protected files to a PostgreSQL or MySQL database running on the same or a completely different underlying architecture. This includes both cloud and on-premise solutions, and it is not required that the source database is the same engine as the destination database, which means this plugin can be used as a migration tool for the protected data.

General Features

Below, there is a list of general features this plugin offers:

• Backup and restore of Amazon RDS instances

• Backup and restore of Amazon RDS clusters

• Amazon AWS API-based backups

• Automatic multi-threaded download or upload operations

• Network resiliency mechanisms

• Instances and volume discovery capabilities

• List instances and snapshots through query or list commands

• Auto-generation capabilities when combined with the Scan Plugin

• Restore data to Amazon RDS from the previously generated snapshots in the cloud

  • To the original location (region, availability zone, etc.)

  • To a different location (region, availability zone, etc.)

  • With the same original instance or cluster configuration

  • With a different instance or cluster configuration

  • Point-in-Time recovery

• Restore data from snapshots done during the backup or from any existing identifier

• Restore data to a running database engine from the data stored in Apache Parquet format

• Full & Incremental backup levels

  • Snapshots are Incremental backups by nature in Amazon RDS, as they are based in a non-visible EBS layer

  • Exports to S3 include always all data from the selected tables

• Advanced instance selection for backup, filtering by any parameter

• Ability to select whether to export data to S3

• Ability to select whether to download data from S3

• Ability to select if all tables or just part of them should be exported to S3 using Apache Parquet format

• Automatic snapshot cleanup before and after backups based on the specified retention

• Configurable automatic S3 bucket cleanup after backups

See also

Previous articles:

• Scope

Next articles:

• Architecture

• Installation

• Configuration

• Best Practices

• Operations

• Limitations

• Troubleshooting

Go back to: Amazon RDS Plugin.