Enabling Backup Archiving

When you combine multiple types of schedules, you can enable the archiving mechanism to instruct the backup appliance to store backed-up data in the secure, low-cost and long-term S3 Glacier Flexible Retrieval and S3 Glacier Deep Archive storage classes. The mechanism is the most useful in the following cases:

• Your data retention policy requires that you keep rarely accessed data in an archive.
• You want to reduce data-at-rest costs and to save space in the high-cost, short-term S3 standard storage class.

With backup archiving, the backup appliance can retain backup files created according to a daily, weekly or monthly schedule for longer periods of time:

• To enable monthly archiving, you must configure a daily or a weekly schedule (or both).
• To enable yearly archiving, you must configure a daily, a weekly or a monthly schedule (or all three).

For the backup appliance to use the archiving mechanism, there must be specified at least 2 different schedules: one schedule will control the regular creation of backup files, while another schedule will control the process of copying backup files to an archive backup repository. Backup chains created according to these two schedules will be completely different — for more information, see RDS Backup Chain and Archive Backup Chain.

Consider the following example. You want a backup policy to create image-level backups of your critical workloads once a week, to keep the backed-up data in a standard backup repository for 3 weeks, and also to keep backups created once in 2 months in an archive backup repository for a year. In this case, you create 2 schedules when configuring the backup policy settings — weekly and monthly:

1. In the policy target settings, you set the Enable backups toggle to On, select a backup repository that will store standard backup files, and select an archive backup repository that will store archived data.

2. In the weekly scheduling settings, you select hours and days when backups will be created (for example, 7:00 AM, Monday), and specify a number of days for which the backup appliance will retain backups (for example, 21 days).

The backup appliance will propagate these settings to the archive schedule (which is the monthly schedule in our example).

3. In the monthly scheduling settings, you enable the archiving mechanism by setting the Send backups to archive toggle to On, specify when the backup appliance will create archive backup files, and choose for how long you want to keep the created backups in the archive backup repository.

For example, January, March, May, July, September, November, 12 months and First Monday.

According to the specified scheduling settings, the backup appliance will create image-level backups in the following way:

1. On the first Monday of February, a backup session will start at 7:00 AM to create the first restore point in the standard backup chain. The backup appliance will store this restore point as a full backup file in the backup repository.
2. On the second and third Mondays of February, the backup appliance will create restore points at 7:00 AM and add them to the standard backup chain as a full backup file in the backup repository.

3. On the fourth Monday of February, the backup appliance will create a new restore point at 7:00 AM. By the moment the backup session completes, the earliest restore point in the standard backup chain will get older than the specified retention limit. That is why the backup appliance will remove from the chain the restore point created on the first Monday.

For more information on how backup appliances transform standard backup chains, see RDS Backup Retention.

4. On the first Monday of March, a backup session will start at 7:00 AM to create another restore point in the standard backup chain. At the same time, the earliest restore point in the standard backup chain will get older than the specified retention limit again. That is why the backup appliance will remove from the chain the restore point created on the second Monday.

After the backup session completes, an archive session will create a restore point with all data from the standard backup chain. The backup appliance will copy this restore point as a full archive backup file to the archive backup repository.

5. Up to May, the backup appliance will continue adding new restore points to the standard backup chain and deleting outdated backup files from the backup repository, according to the specified weekly scheduling settings.

On the first Monday of May, an archive session will create a restore point. The backup appliance will copy this restore point as an incremental archive backup file to the archive backup repository.

6. Up to the first Monday of March of the next year, the backup appliance will continue adding new restore points to the standard backup chain and deleting outdated backup files from the backup repository, according to the specified weekly scheduling settings. The backup appliance will also continue adding new restore points to the archive backup chain, according to the specified monthly settings.

By the moment the archive session completes, the earliest restore point in the archive backup chain will get older than the specified retention limit. That is why the backup appliance remove from the chain the restore point created on the first Monday of March of the previous year.

For more information on how the backup appliance transforms archive backup chains, see Retention Policy for Archived Backups.