EPI Data Centre Foundation flashcards

31 cards. Read the term and the definition here. The deck at the top of this page is optional.

Concurrently maintainable

A Tier III characteristic: any piece of power or cooling equipment on one distribution path can be removed for maintenance while the IT load continues to be served from the other path.

Fault tolerant topology

A Tier IV design in which a failure of any single component or distribution path does not interrupt the IT load, because independent compartments each carry full capacity.

2N redundancy

Two complete independent systems, each sized to carry the full critical load, so one system can fail or be maintained without affecting the IT equipment.

PUE formula

Power Usage Effectiveness equals total facility energy divided by IT equipment energy over the same period. A value of 1.0 is theoretical perfection; lower values indicate better efficiency.

Hot and cold aisle

A layout where equipment fronts face each other in cold aisles fed by cool supply air, and rears face each other in hot aisles returning warm air to the cooling units.

Heat load conversion

One kilowatt of electrical IT load converts to approximately 3,412 BTU per hour of heat, which cooling capacity must remove to keep temperatures stable.

UPS sizing rule

UPS capacity in kVA equals critical load in kW divided by the load power factor, then multiplied by a design margin and redundancy factor.

Cable pathway diversity

Redundant network links must traverse physically separate routes, because two cables in the same tray or duct share the same single point of failure.

Method of Procedure

A task-specific document listing exact steps, prerequisites, safety controls, rollback actions and verification checks for a planned activity on critical infrastructure.

Standard Operating Procedure

A documented, pre-approved sequence for routine recurring operations, such as generator checks or tape rotation, that does not need case-by-case engineering review.

Emergency Operating Procedure

A pre-written response guide for unplanned events, such as UPS failure or cooling loss, defining immediate actions, escalation contacts and load-shedding priorities.

Change management trigger

Any work that can affect a live critical path or a documented configuration requires a change record, risk assessment and approval before execution.

Load bank test

A test applying controlled resistive or resistive-reactive load to a generator or UPS to prove it can carry rated capacity, not merely start and run unloaded.

Shift handover record

A documented transfer of responsibility covering open work orders, abnormal conditions, active permits, bypassed alarms and outstanding risks between operational shifts.

Capacity threshold flag

A predefined utilisation level for power, cooling, space or ports that triggers review before the resource becomes a hard constraint on new deployments.

Incident severity

A classification based primarily on customer impact and loss of redundancy, used to drive escalation speed and communication, rather than on alarm count.

ISO/IEC 27001

The international standard for an Information Security Management System, requiring risk assessment, documented controls, internal audit and management review for certification.

ISO/IEC 22237 series

An international standard covering data centre facility design, operation and protection, structured around availability, security and energy efficiency classifications.

TIA-942

A telecommunications infrastructure standard for data centres that defines cabling, topology and Rated-1 to Rated-4 availability levels for facility infrastructure.

Nonconformity

An audit finding where a documented requirement, such as a standard clause or internal policy, is demonstrably not met, requiring corrective action and follow-up verification.

Audit evidence quality

Evidence strength ranks system-generated records and logs highest, reviewed documents and screenshots next, and verbal statements lowest, because statements are not verifiable.

Segregation of duties

A control ensuring the person who performs a change is not also the sole approver or sole verifier, reducing error and fraud risk.

Document control

A process giving each procedure a version, owner, approval date and review cycle, and withdrawing superseded copies so only current documents are used.

Physical access audit

Reconciling access logs against current authorisation lists, checking visitor escort records, CCTV retention and door alarm tests to confirm physical security controls operate.

Risk treatment options

The four responses to an identified risk: avoid the activity, reduce likelihood or impact, transfer it via insurance or contract, or accept it with documented owner approval.

Risk register fields

Minimum content for each entry: description, owner, likelihood, impact, rating, treatment plan, target date and residual rating after treatment.

RTO and RPO

Recovery Time Objective is the maximum acceptable outage duration; Recovery Point Objective is the maximum acceptable data loss measured in time.

RACI matrix

A responsibility model assigning each activity as Responsible (does the work), Accountable (owns the outcome), Consulted (gives input) or Informed (kept updated).

Key risk indicator

A metric tracking exposure to future adverse events, such as count of overdue critical maintenance or open high-severity audit findings, rather than past performance.

Vendor right-to-audit

A contract clause allowing the client to inspect a supplier's controls and records, supporting assurance over outsourced data centre or maintenance services.

Competence record

Evidence that a person is capable for a task, combining training, formal assessment and supervised practice, rather than attendance certificates alone.