Backing up data is an essential practice for protecting against data loss. However, where you store the backup is just as important as doing the backup itself. Storing a backup in the same location as the original data defeats the purpose of having a backup. If a failure or disaster occurs that damages or destroys the original data, the backup will be affected as well if it is stored in the same place.
What is the purpose of a backup?
A backup provides a secondary copy of data that can be used to restore the original after data loss. The backup allows you to recover critical data and resume operations quickly.
Why is a backup stored separately from the original data?
Storing the backup separately isolates it from failures or disasters affecting the original data. This ensures the backup remains intact and can be used for recovery if needed.
What types of failures can damage both original and backup data?
On-site disasters like fires, floods, electrical surges, and hardware failures can damage collocated originals and backups. Human errors like accidental deletion can also affect both copies.
Where should backups be stored?
Backups should be stored off-site or in the cloud. This provides physical and logical separation from the original data. Popular options include external hard drives, tape drives, cloud storage, and dedicated disaster recovery sites.
Risks of Collocating Original and Backup Data
Storing backups alongside original data introduces risks that can compromise both copies. If a failure or disaster occurs, you can potentially lose both the original and backup data. This defeats the redundancy and recoverability that backups are designed to provide. Key risks include:
Fires, floods, electrical surges, and other disasters localized to the site of the original data can damage or destroy collocated backups. For example, an electrical surge could fry both original data on servers and backups on NAS devices if they are in the same physical location. Physical proximity leads to a single point of failure.
Well-intentioned but accidental human errors like deleting data can affect both original and collocated backup data. For example, an admin could run a delete or reformat command that wipes out original data and backups on the same SAN or NAS device. Without physical separation, both are vulnerable.
Failures of storage hardware like drives, controllers, and RAID arrays can corrupt or lose data. If original and backup data reside on the same hardware, the failure impacts both copies. Hardware redundancy within a single site does not protect against localized disasters.
Malware, ransomware, or insider threats can potentially destroy both original and backup data if they are in the same place. By storing backups separately, only one copy can be affected by malicious actions.
Human Maintenance Errors
Accidental mishaps during routine maintenance like upgrading storage hardware or reconfiguring network gear can impact local data and backups. Physical separation minimizes shared dependencies and risks.
Benefits of Separating Original and Backup Data
Maintaining physical and logical separation between original data and backups provides key benefits:
Isolation from Local Failures and Disasters
Isolating backups in a different location protects them from on-site disasters, hardware failures, and human errors affecting local systems. This separation is critical for recoverability when the unexpected occurs.
Data Redundancy and Availability
Separation ensures original data and backups do not have a single point of failure. If the original data is lost, the backup remains available to restore operations. This provides greater redundancy and resilience.
Recovery Point Objectives
Backups that are separated from original data can meet recovery point objectives for recent point-in-time restores. Backup schedules can run as frequently as needed without affecting local performance.
Compliance and Security
Off-site backups may improve compliance with regulatory mandates for backup strategies. Data separation also makes it harder for malicious actors to access and destroy both original and backup data.
Media Rotation and Lifecycle Management
Storing backups off-site or in the cloud makes it easier to implement policies for rotating and replacing backup media. This maintains backup integrity over time.
Recommended Backup Locations
When choosing a separate location for backups, key options to consider include:
External hard drives provide a simple way to maintain on-site backups that are isolated from primary storage. However, they are still vulnerable to local disasters. Rotation is required to guard against media failure over time.
Maintaining backups at an alternate facility protects against site disasters. This requires manually transporting and managing media. Costs scale with backup size and frequency.
Public cloud services like S3, Glacier, and Azure Backup offer scalable, automated off-site storage. Backups are transmitted over the network, so bandwidth must be considered.
Dedicated Disaster Recovery Sites
For the highest reliability, a dedicated DR site with replicated data provides low RTO restores. This has a higher cost but the highest availability.
Regularly cycling backup tapes to an off-site vault protects against site disasters while managing costs. This requires manual effort and has slower recovery times.
Keeping backups separate from original data is critical to providing meaningful recoverability when failures and disasters occur. On-site backups have a single point of failure and are vulnerable to any incident affecting local systems and data. Storing backups off-site or in the cloud isolates them from local disasters, hardware failures, human errors, and malicious actions. This separation enables much faster and more reliable recovery, avoiding extended downtime and data loss. With backups playing an essential role in modern business continuity, choosing the right backup location is a key consideration when designing a backup strategy.