Skip to main content
Intelligent Security Governance — 340+ Frameworks Powered by Ayurak AI
Blog/Compliance Guides

Cyber Centre Network Security Zones Guidance (ITSP.80.022, ex-ITSG-22)

What the Canadian Centre for Cyber Security's network security zoning guidance says, who uses it, and how to apply zone-based design and evidence it in assessments.

Cyber Centre Network Security Zones Guidance (ITSP.80.022, ex-ITSG-22)

Many Canadian organizations still search for "ITSG-22", the Communications Security Establishment's older document on network security zones. The Canadian Centre for Cyber Security (the Cyber Centre) now publishes this guidance as ITSP.80.022, and it remains a useful reference for segmenting networks by trust level. This article explains what it is, who uses it, and how to apply it in design and assessment work.

What it is

ITSP.80.022 is a Cyber Centre publication that sets out baseline security requirements for network security zones. It carries forward the content of ITSG-22, "Baseline Security Requirements for Network Security Zones in the Government of Canada", first issued by the Communications Security Establishment in the late 2000s. The Cyber Centre is part of the CSE and issues IT security guidance for the federal government and, more widely, for Canadian organizations. The "ITSG" series was the older naming for guidance, and "ITSP" is the newer naming for publications. Exact edition dates and version numbers should be checked on the Cyber Centre site, because the document has been renumbered and revised over time.

It is guidance, not a statute, and not a certifiable standard. There is no certificate you receive for "complying with ITSP.80.022". Its purpose is to give a shared vocabulary and a minimum set of expectations for dividing a network into zones, each with defined security characteristics, and for controlling how traffic moves between them.

It is usually read together with the companion design guidance on placing services within zones (published as ITSG-38; check the Cyber Centre site for its current status), and with the Cyber Centre's IT security risk management guidance (ITSG-33), whose control catalogue the zoning requirements are designed to support.

Who uses it

  • Federal departments and agencies. The guidance was written for the Government of Canada. Its force there comes through Treasury Board security policy instruments and departmental security assessment processes, rather than through the document itself. Confirm the current policy references with your departmental security team.
  • Contractors and service providers to government. Cloud and managed service providers handling government workloads are often asked to show architecture consistent with the zoning model.
  • Provinces, municipalities, critical infrastructure and other Canadian organizations. It is voluntary for them, but it is a well-known, publicly available reference for segmentation and is often adopted as a design baseline.

There are no size or revenue thresholds. If you are not in a government context, treat it as a recommended practice and not as an obligation.

What it defines and recommends

The core idea is to group systems into zones according to the sensitivity of the information and the exposure of the services, and then to enforce controls at the boundaries and within each zone. The ITSG-22 model is commonly described with zones that run from the uncontrolled public side to the most protected internal side; the main ones are listed below, and the guidance also describes further zones such as a Highly Restricted Zone and a Restricted Extranet Zone:

  • Public Zone (PZ): the Internet and other networks outside your control.
  • Public Access Zone (PAZ): the interface between the public side and your internal networks, where externally facing services sit.
  • Operations Zone (OZ): the normal internal environment for day-to-day business systems and users.
  • Restricted Zone (RZ): a more protected zone for sensitive systems and data, with tighter access and monitoring.

Treat the exact zone names and any additional zone types as something to verify against the current edition. The key control areas in the guidance include:

  • Zone definition and characteristics. Each zone has expected properties, such as who may be in it, what is placed in it, and the level of assurance required of the controls around it.
  • Boundary protection and traffic control. Traffic between zones is denied by default and permitted only for documented, justified flows, enforced by firewalls or equivalent devices.
  • Restricting direct paths. Systems in less trusted zones should not connect directly to more trusted ones. Flows are mediated, for example through proxies or services placed in an intermediate zone.
  • Management and administration paths. Administrative access is separated from user and service traffic and controlled tightly.
  • Monitoring and logging. Zone boundaries are points to log, inspect and alert on traffic.
  • Alignment with risk management. Zone placement and controls are tied to the system's categorization and to the control profile selected under the wider risk management process.

The design guidance goes further than the baseline requirements and covers where to place common services such as web, DNS, email, remote access and databases.

How it is used in assessment and audit work

Because there is no certification scheme, the guidance is applied through the assessments you are already subject to:

  • Security assessment and authorization. In government, assessors review network architecture diagrams, firewall rule sets and zone placement as part of authorizing a system to operate.
  • Contractual and procurement reviews. Customers may ask for architecture evidence that maps to the zoning model.
  • Internal audit and architecture review. Many organizations use the zone model as a yardstick for segmentation reviews, penetration test scoping and cloud landing zone design.

Assessors typically look for a current network diagram showing zones, a documented rationale for each permitted flow, rule set reviews, and logs showing the boundaries are monitored. No penalties attach to the document itself. Consequences of non-alignment arise through policy, authorization decisions or contract terms.

How to apply it: a practical path

  1. Confirm your context. Establish whether the guidance is binding through government policy or contract, or is a voluntary baseline for you. Download the current edition and the companion design guidance.
  2. Inventory systems and data flows. List applications, hosts, cloud accounts and networks, and document how they communicate. Include cloud and SaaS connections.
  3. Categorize by sensitivity and exposure. Classify each system by the sensitivity of its data and by whether it is externally reachable.
  4. Assign systems to zones. Map each system to a zone and note exceptions, such as legacy systems that cannot be moved yet.
  5. Define the permitted flows. Write down which inter-zone flows are allowed, the reason, and the owner. Default to deny for everything else.
  6. Implement and harden boundaries. Configure firewalls, security groups, network ACLs, proxies and segmentation in cloud and on premises. Separate management traffic.
  7. Log and monitor. Send boundary and zone logs to central monitoring, with alerts for unexpected cross-zone traffic.
  8. Test and review. Validate rule sets, run segmentation tests, and review rules on a schedule and after changes.
  9. Keep evidence ready. Maintain current diagrams, rule reviews, change records and monitoring output so assessors can see the model in operation.

How it maps to other frameworks

The overlap is real, but these are overlaps, not equivalences, because ITSP.80.022 is a design guidance document and the others are control frameworks or standards.

  • ITSG-33. The closest relationship. Zoning supports the system and communications protection and access control families in the Cyber Centre's control catalogue, which is itself modelled on NIST SP 800-53.
  • NIST SP 800-53 and the NIST CSF. Boundary protection, network segmentation and monitoring correspond to controls in the System and Communications Protection family and to the Protect and Detect functions of the CSF.
  • ISO/IEC 27001. Annex A controls on network security and segregation of networks address the same objective.
  • SOC 2. Logical access and system operations criteria are typically evidenced by segmentation and firewall configurations, though SOC 2 does not prescribe a zone model.
  • PCI DSS. Its network segmentation and scoping requirements overlap in practice with zone-based design for cardholder data environments.

FAQs

Is ITSG-22 withdrawn?

ITSG-22 is the older identifier. The Cyber Centre now publishes the guidance as ITSP.80.022. Use the current publication for new work, and note that older assessments or contracts may still cite ITSG-22 by name.

Can we be certified against ITSP.80.022?

No. It is guidance with no certification or attestation scheme of its own. You show alignment through architecture documentation, assessments and technical evidence.

Does it apply to cloud environments?

The zoning principles apply, but the original model was written with traditional networks in mind. In cloud you implement zones with accounts, virtual networks, security groups and policy controls. Government cloud use is also subject to separate Cyber Centre and Treasury Board guidance.

Is it mandatory for private companies?

Not by itself. It becomes relevant if a contract, a government customer or your sector regulator points to it. Otherwise it is a voluntary reference for segmentation.

How does it relate to zero trust?

Zone-based design is a perimeter and segmentation model, while zero trust emphasizes per-request verification of identity and context. Many organizations use zones as a foundation and layer identity-centric controls on top.

How Aristiun helps

Aristiun is the AI-powered compliance automation and threat modeling platform, powered by Ayurak AI. For work aligned to network zoning guidance, it helps you collect evidence and keep it current rather than rebuilding it for each assessment. You can browse the compliance frameworks it covers.

  • Automated evidence collection: gather configuration evidence from connected systems such as AWS, Azure, GCP and GitHub, without manual screenshots or spreadsheets.
  • Cross-framework mapping: map controls across 340+ frameworks, so a single evidence base can address multiple compliance requirements simultaneously.
  • Real-time control monitoring: get alerts when controls drift out of compliance, so you can remediate before auditors find issues.
  • Automated threat modeling: generate threat models from code, IaC and architecture diagrams using methodologies such as STRIDE, PASTA, LINDDUN and MAESTRO.

Found this insightful?

Share it with your security team or save it for later.

Get Started

Ready to secure your organization?

See how Aristiun automates compliance and strengthens your security posture across 340+ frameworks.