ISO 27001 Redundancy of Information Processing Facilities
ISO 27001 Annex A 8.14 Redundancy of information processing facilities is an ISO 27001 control that requires an organisation to implement information processing facilities with redundancy built in that is sufficient enough to meet availability requirements.
Table of contents
- ISO 27001 Redundancy of Information Processing Facilities
- Key Takeaways
- Explanation
- Requirement
- Audit Focus
- FREE Training Video
- Identify Requirements
- Design and Implement Redundancy
- Implement Alerts
- Cloud Computing
- Testing
- Single Point of Failure (SPOF)
- Business Continuity Templates
- How to implement it
- How to comply
- What will an auditor check?
- ISO 27001 Templates
- Top 3 mistakes and how to avoid them
- Applicability across different business models
- FAQ
- Related ISO 27001 Controls
Key Takeaways
ISO 27001 Annex A 8.14 requires organizations to implement redundancy in their IT systems to ensure they remain operational during a failure. It’s not just about “having a backup”; it’s about having standby hardware, software, or network connections that can take over immediately if the primary system fails. The goal is to meet the availability requirements defined by the business.
Purpose
ISO 27001 Annex A 8.14 is preventive control that ensures the continuous operation of information processing facilities.
Definition
The ISO 27001 standard defines ISO 27001 Annex A 8.14 as:
Information processing facilities should be implemented with redundancy sufficient to meet availability
ISO27001:2022 Annex A 8.14 Redundancy of Information Processing Facilities
Explanation
ISO 27001 Annex A 8.14 is a security control that mandates the implementation of redundancy in information processing facilities to ensure sufficient availability. It requires organizations to identify critical systems and deploy failover mechanisms (such as duplicate servers, network links, or power supplies) to guarantee continuous operation during disruptions and meet defined recovery time objectives.
Requirement
- Eliminate Single Points of Failure (SPOF): You must identify any single component that, if it fails, would take down your entire service (e.g., a single internet line, one power supply, or a single database server).
- Technical Duplication: Depending on your risk, this may include dual power supplies, mirrored databases, load balancers, or redundant internet service providers (ISPs).
- Automatic vs. Manual Failover: You must define how the system switches to the redundant component. For critical systems, this should be automatic; for less critical ones, a documented manual process is acceptable.
- Testing: Redundancy is a theory until it is tested. You must regularly simulate a failure (e.g., pulling a power cable or failing over a database) to prove your standby systems actually work.
Audit Focus
- SPOF Analysis: “Show me your list of critical systems. Do any of them rely on a single server or internet link?”
- Evidence of Uptime: “Show me the report from your last failover test. Did the redundant system take over within the agreed timeframe?”
- Cloud Leveraging: If you are in the cloud (AWS/Azure), they will check if you are using “Multi-AZ” (Availability Zones) to ensure redundancy across different physical data centers.
FREE Training Video
In this free training video you will learn How to implement ISO 27001 Redundancy of Information Processing Facilities (Annex A 8.14 )
Identify Requirements
Identify what requirements you have for maintaining availability so that you can understand what you need to implement.
The best way to identify the redundancy requirements is to conduct a business impact assessment (BIA). The BIA will identify and prioritise your critical systems and data and will provide you with time scales for how quickly they should be recovered. This in turn will inform your approach to redundancy.
Design and Implement Redundancy
This is where you will need the guidance of your subject matter experts in the technologies that you have. Usually there is an element of duplication that meets redundancy requirements and then implementing the processes and procedures to activate those redundant components.
Implement Alerts
You will want to have alerts in place that notify you when any element that is covered under redundancy fails so that you can respond and run the processes and procedures you have in place to recover.
Cloud Computing
Cloud computing allows you various advantages including the the potential to have multiple live versions of information processing facilities, multiple separate physical locations, automatic failover, load balancing and more. Work with the subject matter experts to see what you can leverage to meet your requirements.
Testing
It goes without saying that you should test and evidence your testing for activating redundancy. This is usually as part of your business continuity and disaster recovery testing.
Single Point of Failure (SPOF)
An example checklist for SPOF:
- Do you have two ISPs?
- Do servers have dual power supplies?
- Is there a UPS battery backup?
- Is the database clustered?
Business Continuity Templates
The business continuity templates set out the approach to redundancy of information processing facilities.
How to implement it
Implementing redundancy for information processing facilities is a critical requirement for ensuring continuous service availability and operational resilience against hardware failures or regional outages. By following these technical implementation steps, your organisation can satisfy ISO 27001 Annex A 8.14 requirements and mitigate the risk of single points of failure within your infrastructure.
1. Formalise Availability Requirements and RTO/RPO Metrics
- Identify all critical business processes and document the required Recovery Time Objectives (RTO) and Recovery Point Objectives (RPO) for each.
- Establish a formal Business Impact Analysis (BIA) to determine the level of redundancy needed (e.g. Active-Active vs Active-Passive) based on risk appetite.
- Result: A documented technical roadmap that ensures redundancy investments are aligned with business continuity needs.
2. Provision Multi-Zone and Multi-Region Infrastructure
- Deploy application workloads across multiple Availability Zones (AZs) or distinct geographic regions to protect against localised data centre failures.
- Utilise Virtual Private Cloud (VPC) peering or dedicated interconnects to ensure low-latency synchronisation between redundant processing nodes.
- Result: Increased fault tolerance by eliminating geographic single points of failure for critical information systems.
3. Implement Hardware and Network Redundancy
- Provision redundant hardware components, including dual power supplies, RAID configurations for storage, and multiple Network Interface Cards (NICs).
- Configure diverse network paths via different Internet Service Providers (ISPs) to maintain connectivity during external infrastructure outages.
- Result: Enhanced physical resilience that allows individual component failures to occur without impacting overall service delivery.
4. Configure Automated Load Balancing and Failover
- Deploy Application Load Balancers (ALB) to distribute traffic across redundant server instances and perform continuous health checks.
- Enable automated failover mechanisms for databases using synchronous replication and Multi-AZ deployments to ensure zero data loss during a switchover.
- Result: Seamless transition of user traffic to healthy nodes during a failure, maintaining high availability without manual intervention.
5. Restrict Redundancy Management via IAM and MFA
- Apply the Principle of Least Privilege by assigning specific Identity and Access Management (IAM) roles to administrators managing failover configurations.
- Mandate Multi-Factor Authentication (MFA) for any changes to DNS records or load balancer settings to prevent malicious redirection of traffic.
- Result: Protection of the redundant infrastructure from unauthorised modifications or accidental de-provisioning by privileged users.
6. Execute Regular Redundancy Testing and Audits
- Formalise a Rules of Engagement (ROE) document for “Chaos Engineering” or failover testing to verify that redundant systems trigger correctly under stress.
- Conduct quarterly technical audits of synchronisation logs to ensure that data mirrors remain consistent across all redundant processing facilities.
- Result: Verifiable proof for ISO 27001 auditors that redundancy controls are functional and capable of meeting defined availability targets.
Hello. I am Stuart Barker.
CEO here at High Table: The Compliance Agency
If you want help by the hour, internal audit or consulting support …
How to comply
To comply with ISO 27001 Annex A 8.14 you are going to implement the ‘how’ to the ‘what’ the control is expecting. In short measure you are going to:
- Understand and record the legal, regulatory and contractual requirements you have for data
- Conduct a risk assessment
- Based on the legal, regulatory, contractual requirements and the risk assessment you will implement a redundancy solution
- Document and implement your processes and technical implementations for redundancy
- Check that the controls are working by conducting internal audits
What will an auditor check?
The audit is going to check a number of areas. Lets go through the main ones
- That you have documentation: What this means is that you need to show that you have documented your legal, regulatory and contractual requirements for information and that you have taken this into account when building your information processing facilities redundancy. Where data protection laws exist that you have documented what those laws are and what those requirements are. That you have an information classification scheme and a topic specific policy for access control and that you have documented your information redundancy taking all of this into account.
- That you have have implemented redundancy appropriately: They will look at systems to seek evidence of information processing facilities redundancy, testing and recovery. They want to see evidence of tests, the results of tests and any continual improvement you conducted as a result of those tests.
- That you have conducted internal audits: The audit will want to see that you have tested the controls and evidenced that they are operating. This is usually in the form of the required internal audits. They will check the records and outputs of those internal audits.
ISO 27001 Templates
Top 3 mistakes and how to avoid them
In my experience, the top 3 mistakes people make for ISO 27001 Annex A 8.14 Redundancy of information processing facilities are
- You have not tested: This is a common mistake we see. That you have not tested that you can recover and activate your redundancy solution. Sometimes you did a recovery test but it was a long time ago, or it was a partial recovery and therefore you have no actual evidence that your redundancy solution can be implemented to a point the organisation is operational again within the time frames and to the point in time that was agreed.
- You don’t know your legal obligations: This is a massive mistake that we see, where people assume ISO 27001 is just information security and forget that it also checks that appropriate laws are being followed, and in particular data protection laws. Cost saving by not having a data protection expert or ignoring data protection law entirely is a common mistake we see people make when cutting corners and saving costs. Duplicating data and having redundancy, in particular personal information, can get you in a lot of hot water depending how you implement it.
- Your document and version control is wrong: Keeping your document version control up to date, making sure that version numbers match where used, having a review evidenced in the last 12 months, having documents that have no comments in are all good practices.
Applicability across different business models
| Business Type | Applicability | Examples of Control Implementation |
|---|---|---|
| Small Businesses | Focuses on protecting physical office operations and basic internet connectivity. The goal is to ensure that a single hardware failure or power cut doesn’t stop the business from operating. |
|
| Tech Startups | Fundamental for maintaining customer-facing service availability. Compliance involves leveraging cloud-native features to automate failover and ensure 99.9%+ uptime. |
|
| AI Companies | Vital for high-performance computing (HPC) and data ingestion. Focus is on ensuring that massive model training jobs aren’t lost due to single-node failures. |
|
FAQ
Redundancy ensures real-time business continuity, while Backup ensures data recovery after a catastrophic loss. While they are complementary, they serve different functions in an ISMS.
Redundancy (Annex A 8.14): Keeps systems running during an incident (e.g., a secondary server takes over immediately if the primary fails).
Backup (Annex A 8.13): Restores data after an incident (e.g., restoring a database from last night’s copy after corruption).
Speed: Redundancy offers near-zero downtime; Backups have a Recovery Time Objective (RTO).
No, redundancy is only required for information processing facilities where availability is critical to the organization. The standard explicitly states redundancy must be “sufficient to meet availability requirements.”
Step 1: Conduct a Business Impact Analysis (BIA) to identify critical assets.
Step 2: Assign redundancy only to systems with a low tolerance for downtime.
Cost-Efficiency: You do not need expensive active-active clusters for non-critical internal tools.
Common examples include duplicating power, network connectivity, and storage hardware to prevent a single failure from stopping operations. Implementation varies based on whether you are on-premise or cloud-based.
Power: Uninterruptible Power Supplies (UPS) and backup generators.
Network: Dual Internet Service Providers (ISPs) entering the building via different physical routes.
Compute: Load balancers distributing traffic across multiple servers.
Storage: RAID (Redundant Array of Independent Disks) or mirrored databases.
Cloud computing simplifies compliance by offering built-in redundancy features, but the responsibility remains with the organization to configure them correctly. Simply moving to the cloud does not guarantee redundancy.
Availability Zones (AZ): You must configure services to run across multiple physical data centers (Multi-AZ) within a region.
Geo-Redundancy: For high-risk systems, replicate data across different geographic regions.
SLA Gaps: Review your provider’s Service Level Agreement (SLA) to ensure their guaranteed uptime matches your business needs.
Yes, untested redundancy is viewed as non-compliant because there is no assurance it will work when needed. Testing ensures that failover mechanisms trigger correctly and within the required timeframes.
Frequency: Test annually or after significant infrastructure changes.
Types of Tests: Tabletop exercises, live failover drills, and restoration tests.
Record Keeping: Always document the outcome of the test and any remedial actions taken if the failover failed.
Related ISO 27001 Controls
- ISO 27001 Annex A 5.30 ICT Readiness for business continuity
- ISO 27001 Annex A 7.3 Securing Offices, Rooms And Facilities
- ISO 27001 Annex A 7.11 Supporting Utilities
- ISO 27001 Annex A 8.29 Security Testing in Development and Acceptance
- ISO 27001 Annex A 8.34 Protection of Information Systems During Audit Testing
- ISO 27001 Annex A 8.33 Test Information
About the author
