Synthetic full backup is a backup technique that creates a new full backup by consolidating and combining a baseline full backup with subsequent incremental backups, without requiring a new full data capture operation from source systems.
A synthetic full backup addresses a fundamental challenge with incremental backup strategies. Incremental backups are storage-efficient, creating chains of backups where each incremental depends on previous backups. However, recovery from old incremental backups requires accessing and applying multiple backup sets sequentially, and removing older incremental backups from retention can break chains for backups still under retention. Synthetic full backups solve this by consolidating incremental chains into new full backups that reference only the consolidated result, rather than requiring new full backup operations that consume source system resources.
Why Synthetic Full Backup Transforms Incremental Backup Economics
For IT directors managing incremental backup strategies protecting large datasets, synthetic full backup represents a transformative optimization. Without synthetic capability, maintaining frequent full backup baselines for clean recovery points requires actual full backup operations that consume network bandwidth and backup processing resources. With synthetic full backup, organizations can create new full backup baselines simply by consolidating existing incremental backups, consuming only backup storage processing without impacting source systems.
Consider a 50TB database with weekly full backups and daily incremental backups. After four weeks, the incremental chain spans a full backup plus 28 incremental backups. Recovery to a point from three weeks ago requires restoring that three-week-old full backup plus all incremental backups since then. A synthetic full backup consolidates the four-week chain into a new full backup without requiring a fresh 50TB capture from the database. Subsequent incremental backups then reference the new synthetic full backup, resetting the incremental chain.
The resource conservation is substantial. A new full backup operation might require 6+ hours of database resources and 50TB+ of network bandwidth. A synthetic full backup consolidation might require 1-2 hours of backup storage resources and no database or network resources. This allows frequent full backup baselines without the operational overhead of actual full backup operations.
How Synthetic Full Backup Works
Synthetic full backup operates in backup storage, not on source systems. The backup server takes an existing full backup (the baseline) and subsequent incremental backups that reference that baseline, and consolidates them into a new full backup. The backup software reads the baseline full backup, sequentially applies incremental backups in order to logically reconstruct what the complete data looked like after the final incremental, and writes this consolidated view as a new full backup.
The process is logical, not physical. The software doesn’t physically re-read all source data and create a new full backup from scratch. Instead, it reads previously captured data from existing backup sets and recombines them into a new full backup structure. Modern backup software can perform this consolidation while the original incremental chain remains available for recovery—creating the synthetic full without affecting existing backups.
The resulting synthetic full backup is operationally identical to a full backup captured directly from source systems. Recovery from a synthetic full backup requires no knowledge of how the full backup was created. Users and recovery processes treat it as a normal full backup.
Synthetic Full vs. Traditional Full Backup Approaches
Traditional full backup captures data directly from source systems. Synthetic full backup consolidates already-captured backup data within storage systems. Organizations typically use weekly actual full backups, daily incremental backups, with synthetic full backups consolidating incremental chains—providing clean recovery baselines without requiring weekly full backup operations.
Storage-wise, synthetic full backups don’t add permanent capacity requirements because they typically replace (consolidate) the incremental chain they’re created from. An organization might retain the synthetic full backup for long-term recovery capability while deleting the incremental chain it was derived from, simplifying retention policies and catalog management.
Scheduling Synthetic Full Backup Operations
Synthetic full backup operations should be scheduled carefully to avoid impacting backup and recovery windows. The consolidation process consumes backup storage I/O and processing resources that might otherwise be available for running incremental backups or serving recovery requests.
Timing typically aligns with incremental backup completion. If daily incremental backups complete by 9 AM, synthetic full backup might be scheduled to begin at 10 AM, allowing incremental completion before synthetic operations consume storage resources. Alternatively, some organizations schedule synthetic backups during lowest activity periods—mid-day—to avoid contention with morning incremental backups or afternoon recovery requests.
For very large backup volumes, synthetic full backup operations might take many hours. A 100TB incremental chain consolidated into a synthetic full backup might require 4-6 hours depending on storage I/O capacity. Organizations should forecast synthetic full completion time and ensure adequate backup window time.
Storage Considerations with Synthetic Full Backup
Synthetic full backup increases temporary storage during consolidation. Well-designed implementations minimize overhead through incremental reading/writing. Long-term storage typically decreases because incremental chains can be deleted after synthetic consolidation, retaining only synthetic full backup for older recovery points.
Verification and Reliability of Synthetic Backups
Synthetic full backups should undergo backup verification as traditional full backups. Regular test restoration to isolated test environments prevents discovery of synthetic backup failures only during actual recovery.