The identification of emanation security controls that need to be implemented for a system at an early stage in the project lifecycle. This can
significantly affect project costs. Costs are invariably greater where changes are necessary once the system had been designed or has been
implemented.
10.7.8.C.01.
ControlSystem Classification(s): All Classifications; Compliance: SHOULD
[CID:2463]
Agencies SHOULD conduct an emanation security threat assessment as early as possible in project lifecycles.
10.7.9.
IT equipment in SECURE areas
10.7.9.R.01.
Rationale
All equipment must conform to applicable industry and government standards, including NZCSI 420; Laboratory Tempest Test Standard for
Equipment in Controlled Environments. Not all equipment within a secure facility in New Zealand requires testing against TEMPEST standards.
10.7.9.C.01.
ControlSystem Classification(s): Confidential, Secret, Top Secret; Compliance: MUST
[CID:2465]
Agencies MUST ensure that IT equipment within secure areas meet industry and government standards relating to electromagnetic
interference/electromagnetic compatibility.
10.8. Network Design, Architecture and IP Address Management
Objective
10.8.1. IP Address architecture, allocation and addressing schemes enable and support system security and data protection.
Context
Scope
10.8.2. This section includes discussion of the principles of separation and segregation as network design and network architecture characteristics. It also
discusses how IP addresses can be used to support these principles for improved security of larger or multi-classification agency systems.
Background
10.8.3. The larger the network, the more difficult it is to protect. A large, unsegmented network presents a large attack surface with greater susceptibility to
the rapid spread and dissemination of system faults, weaknesses, vulnerabilities and attacks. In a non-segmented network, traffic and applications
generally have access to the entire network. If a fault occurs or an attacker gains entry, the fault or attacker can move laterally through the network
causing disruption, network outages and enabling access to critical operational or classified data.
10.8.4. A large network is also more difficult to monitor and control. Segmenting the network limits an attacker’s ability to move through the network by
preventing lateral movement between zones. Segmentation also enhances the ability to detect, monitor, control, isolate and correct faults.
10.8.5. A fundamental construct in the management of risk in networks is that of Trust Zones and Trust Boundaries. A Trust Zone is a zoning construct based
on levels of trust, classification, information asset value and essential information security. A Trust Boundary is the interface between two or more
Trust Zones. Trust Zones use the principles of separation and segregation to manage sensitive information assets and ensure security policies are
consistently applied to all assets in a particular Trust Zone. Refer also to Section 22.1 – Cloud Computing and Section 22.2 – Virtualisation.
Separation and Segregation
10.8.6. Separation and Segregation are determined by system function and the sensitivity of the data the system stores, processes and transmits. One
common example is placing systems that require a connection to the Internet into a demilitarized zone (DMZ) that is separated and segregated
(isolated) from more sensitive systems. Another example is the use of compartments.
10.8.7. Separation and Segregation limits the ability of an intruder to exploit a vulnerability with the intent of elevating privileges to gain access to more
sensitive systems on the internal network. VLANs may be used to further separate systems by controlling access and providing segregation thus giving
additional protection.
10.8.8. Network segmentation is an effective strategy for protecting access to key data assets, and impeding the lateral movement of system faults, threats
and malicious activity. Segregation has the following benefits:
Enhanced performance: with fewer hosts per subnet, there is a reduced signalling and traffic overhead allowing more bandwidth for data
communication.
Improved security: with less signalling traffic going through all network segments, it is more difficult for an attacker to analyse the addressing
scheme and network structure. Failures in one segment are less likely to propagate and more robust access control can be established and
enforced.
10.8.9. Effective segregation also requires:
Specialised knowledge: networks may support many devices with complex policies and rule sets. Support staff must be properly educated and
trained to ensure the network segmentation is maintained.
Administrative effort: changes in infrastructure, such as new applications and new technologies, can extend the time required to make changes
and ensure the integrity of network segments.
Infrastructure Investment: segregation may require more equipment, new equipment with advanced functionalities, or specific software to deal
with multiple segments. These requirements should be considered during budget planning.
ISO 27001 and ISO 27002 implementation recommendations for network segregation
10.8.10. These ISO Standards require the implementation of network segregation. In particular they recommend that groups of information services, users, and
information systems are segregated on networks. Specific recommendations are summarised below:
Divide large networks into separate network domains (segments);
Consider physical and logical segregation;
Define domain perimeters;
Define traffic rules between domains;
114
Version_3.5__January-2022
Select target paragraph3
Connect to a paragraph
Connect to an entity
Disable highlights
Add to table of contents