This document provides best practices for configuring and managing hourly backups for Compute Engine instances. The 1-hour backup frequency is designed for workloads requiring higher data protection granularity. These recommendations help you optimize data protection, manage performance, and set realistic expectations for recovery points.
Combine daily and hourly backup rules
It is strongly recommended to include a daily backup rule alongside a 1-hourly frequency rule in your backup plan. This ensures that even if individual hourly backups are skipped or delayed, a consistent daily recovery point is maintained.
Understand scheduled backup retry logic
Retry behavior varies significantly between Daily and Hourly recurrence types.
Daily Recurrence: If a daily backup job fails to start or complete, it is retried until the end of the configured backup window.
1-Hourly Recurrence: An hourly backup event is retried for 1 hour only, regardless of the length of the overall backup window. This prevents backup jobs from overlapping and causing congestion.
Manage high data churn
High churn can lead to long-running backups that prevent subsequent hourly jobs from starting. To optimize performance and to reduce impact of concurrent backups within a vault, spread high-frequency backups for VMs experiencing high data churn across multiple backup vaults.
Monitor and configure alerts for skipped backups
Configure alerts for skipped backup jobs for proactive notification of protection gaps.
You can set up alerts using Cloud Logging by monitoring for "Backup Plan Violation" events or specific "skipped backup" log entries from the Backup and DR scheduler.
Summary of best practices
The following table summarizes the best practices recommended in this document:
| Topic | Task |
|---|---|
| Backup frequency | Use 1-hour backup frequency judiciously based on workload needs. |
| Backup plan | Combine daily and hourly backup rules for consistent recovery points. |
| Retry logic | Understand the different retry behaviors for daily and hourly backups. |
| Recovery Point Objective | Set realistic RPO expectations that account for churn and performance. |
| High data churn | Reduce backup frequency and distribute workloads for high-churn VMs. |
| Monitoring | Configure alerts for skipped backup jobs to avoid protection gaps. |
What's next
- Learn how to create a backup plan.
- Read about monitoring backup jobs with Cloud Logging.