I85PREP
⌘ K
CDFOM study materials
CDFOM study resources

CDFOM practice questions

One question at a time. Build your CDFOM understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operations team is defining the measurement window for its service level agreement on power availability. Which practice best supports meaningful KPI reporting?

  1. Measure availability only at the utility feed, excluding UPS and generator performance
  2. Report availability only when an incident occurs so the metric reflects real events
  3. Reset the measurement window whenever the facility undergoes maintenance
  4. Report availability monthly and also publish a rolling 12-month figure to smooth seasonal variation Correct answer.

Explanation

KPI reporting is most useful when the measurement window is fixed in advance and both short-term and rolling views are published, so trends and seasonal effects are visible. Reporting only on incidents or resetting the window during maintenance lets the operator choose favourable periods, which undermines comparability and trust.

Question 2Data Centre Operations Management (operational processes, service level and KPI management)

Which metric most directly expresses the reliability of the critical power path in a data centre?

  1. Number of change requests raised per rack
  2. Power usage effectiveness reported monthly
  3. Average time to complete a preventive maintenance work order
  4. Unplanned power-path failures per year affecting the IT load Correct answer.

Explanation

Reliability of the critical power path is captured by counting unplanned failures that actually reach and affect the IT load, usually normalised per year. Efficiency, change volume and maintenance duration describe other aspects of operations and do not measure failure of the power path itself.

Question 3Data Centre Operations Management (operational processes, service level and KPI management)

An operations manager wants a KPI that shows how quickly the team restores service after an unplanned outage. Which measure is most appropriate?

  1. Number of open problem records at month end
  2. Mean time to restore service after an unplanned incident Correct answer.
  3. Percentage of planned maintenance completed on schedule
  4. Mean time between failures for the cooling plant

Explanation

Mean time to restore measures the elapsed time from the start of an unplanned incident to full service restoration, which is exactly the recovery speed question. MTBF, maintenance completion and open problem counts describe failure frequency, planned work and backlog rather than restoration speed.

Question 4Data Centre Operations Management (operational processes, service level and KPI management)

A service level agreement states that critical cooling must be restored within four hours of a failure. Which operational artefact converts this commitment into day-to-day action?

  1. The facility security access matrix
  2. An operational level agreement with internal engineering response targets and escalation paths Correct answer.
  3. A capital replacement plan for chillers
  4. The annual energy efficiency report

Explanation

An operational level agreement translates the customer-facing commitment into internal targets, responsibilities and escalation paths that the engineering team can act on. Capital plans, energy reports and access matrices support other functions and do not operationalise a restoration commitment.

Question 5Data Centre Operations Management (operational processes, service level and KPI management)

Which statement best describes the purpose of a service level objective within data centre operations?

  1. It defines the internal target level of performance used to manage toward the customer commitment Correct answer.
  2. It records the maintenance history of each critical asset
  3. It is the legally binding penalty clause attached to a customer contract
  4. It lists the authorised personnel for each secure zone

Explanation

A service level objective is an internal target set below or at the committed service level so the team can act before the customer commitment is breached. Penalty clauses, asset history and access lists serve contractual, maintenance and security purposes respectively.

Question 6Data Centre Operations Management (operational processes, service level and KPI management)

During a monthly service review, the operations team presents a dashboard showing availability, incident counts and mean time to restore. Which practice keeps this review effective over time?

  1. Change the KPI definitions each month to reflect current priorities
  2. Agree the KPI definitions and targets in advance and review exceptions with corrective actions Correct answer.
  3. Present only the metrics that met their targets
  4. Review only the metrics requested by the largest customer

Explanation

Stable, pre-agreed KPI definitions and targets make month-to-month comparison valid, and focusing the review on exceptions with corrective actions drives improvement. Changing definitions, hiding misses or narrowing the view to one customer destroys the credibility and usefulness of the review.

Question 7Data Centre Operations Management (operational processes, service level and KPI management)

What is the primary purpose of a change management process in a live data centre?

  1. To reduce the number of vendors allowed on site
  2. To assess and control risk to services before modifications are made Correct answer.
  3. To shorten the preventive maintenance schedule
  4. To replace the incident management process

Explanation

Change management exists to evaluate the risk and impact of a proposed modification, plan it, obtain approval and provide rollback, so live services are protected. Vendor control, maintenance scheduling and incident handling are separate processes with different objectives.

Question 8Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator wants to demonstrate that its operations team is prepared for a utility power failure. Which activity provides the most credible evidence for the service review?

  1. Reviewing the manufacturer's generator datasheet
  2. Adding a note to the risk register that the risk is understood
  3. Running a documented load-bank or transfer test and recording the measured outcome against the objective Correct answer.
  4. Confirming that the generator was purchased within the last five years

Explanation

Documented testing that measures actual transfer and load performance against a defined objective is the evidence a service review can rely on. Datasheets, purchase dates and risk register notes describe intent or specification, not demonstrated operational readiness.

Question 9Data Centre Operations Management (operational processes, service level and KPI management)

A data centre reports a PUE of 1.9 while its design target was 1.5. Which action is the most appropriate first step in the operations review?

  1. Ignore the variance because PUE is only a design metric
  2. Publish the figure as a service level breach and invoke penalties
  3. Immediately replace all chillers with a newer model
  4. Analyse the components of the gap, such as part-load efficiency, airflow management and setpoints, before deciding on corrective action Correct answer.

Explanation

A PUE gap should be decomposed into contributing factors so corrective action targets the actual cause rather than assumed equipment failure. Immediate capital replacement, treating an efficiency metric as a contractual breach or dismissing the metric all bypass the diagnostic step that operations management requires.

Question 10Data Centre Operations Management (operational processes, service level and KPI management)

An operator commits to 99.99 percent availability for critical power but measures availability only at the building utility feed. What is the main weakness of this arrangement?

  1. The measurement boundary excludes the UPS and distribution path that actually serve the IT load Correct answer.
  2. Availability cannot be expressed as a percentage at the utility feed
  3. The commitment is too low for a modern data centre
  4. The measurement period is too short to be statistically valid

Explanation

The credibility of an availability commitment depends on measuring at the boundary that matters to the customer, which is the IT load rather than the incoming utility supply. A feed-level measurement can look excellent while UPS, switchgear or distribution faults interrupt the load.

Question 11Data Centre Operations Management (operational processes, service level and KPI management)

A critical incident occurs at 02:00 and service is restored at 05:30. The team spends a further three hours on root cause investigation. Which figure should be reported as mean time to restore for this incident?

  1. Six hours and thirty minutes
  2. Three hours
  3. Two hours
  4. Three hours and thirty minutes Correct answer.

Explanation

Mean time to restore measures elapsed time from incident start to service restoration, which is 02:00 to 05:30, or three hours and thirty minutes. Root cause investigation happens after restoration and is tracked separately, so it must not be added to the restoration figure.

Question 12Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator is negotiating a service level agreement with a colocation customer. Which approach produces the most workable availability commitment?

  1. Define availability precisely, including measurement points, exclusions and credit mechanism, and confirm the design can support it Correct answer.
  2. Leave the availability definition open so it can be interpreted case by case
  3. Commit to the highest figure any competitor advertises
  4. Base the commitment on the best month ever recorded by the facility

Explanation

A workable commitment specifies the measurement points, permitted exclusions and remedy, and is validated against the actual design and operational capability. Aspirational or ambiguous figures create disputes that operations cannot resolve after the fact.

Question 13Data Centre Operations Management (operational processes, service level and KPI management)

An operations team tracks the number of unplanned incidents but finds the figure unchanged after a year of corrective actions. Which additional analysis would best reveal whether the corrective actions are working?

  1. Comparing the incident count with the facility's energy consumption
  2. Reviewing the number of staff trained during the year
  3. Segmenting incidents by cause, severity and affected system to see where reductions occurred Correct answer.
  4. Counting the total number of maintenance work orders completed

Explanation

A flat aggregate count can hide improvements in one cause category offset by growth in another, so segmentation by cause, severity and system shows where corrective action is effective. Work-order volume, training counts and energy data do not attribute incident reduction to specific actions.

Question 14Data Centre Operations Management (operational processes, service level and KPI management)

A data centre wants a KPI to monitor the health of its preventive maintenance programme. Which measure is most informative?

  1. Average age of the maintenance team
  2. Number of maintenance technicians employed
  3. Percentage of scheduled preventive maintenance tasks completed within their due window Correct answer.
  4. Total maintenance spend for the year

Explanation

Completion within the due window shows whether the programme is actually keeping to schedule, which is the core health indicator for preventive maintenance. Spend, headcount and team age describe resources rather than schedule adherence.

Question 15Data Centre Operations Management (operational processes, service level and KPI management)

A colocation provider must report on the timeliness of customer-requested installations. Which measurement approach is most defensible?

  1. Measure from the date the sales team signed the contract
  2. Measure from the date the customer first enquired about the service
  3. Measure from the agreed order acceptance date to the committed delivery date, counting only requests that met the documented prerequisites Correct answer.
  4. Measure only the installations that the provider considers complex

Explanation

A defensible measurement uses an agreed start trigger and a committed end date, and it applies consistent prerequisites so both parties know what is being measured. Measuring from first enquiry, from contract signature or selectively by complexity introduces ambiguity or bias.

Question 16Data Centre Operations Management (operational processes, service level and KPI management)

An operations manager is designing the escalation path for critical incidents. Which design best supports the restoration commitment?

  1. Route all incidents to a single mailbox for review at the next business day
  2. Define time-based and severity-based triggers with named roles and contact routes at each level Correct answer.
  3. Rely on the on-call engineer to decide when to escalate without defined triggers
  4. Escalate to senior management only after the customer complains

Explanation

Defined time-based and severity-based triggers with named roles ensure escalation happens predictably and quickly enough to protect the restoration commitment. Waiting for complaints, leaving escalation to individual judgement or deferring to the next business day all introduce delay.

Question 17Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator wants to reduce the risk that a maintenance activity causes an outage. Which control is most directly relevant?

  1. Reviewing the activity at the next quarterly service review
  2. Adding the activity to the annual capital plan
  3. Increasing the frequency of energy reporting
  4. Requiring a method statement, risk assessment and permit to work before the activity begins Correct answer.

Explanation

A method statement, risk assessment and permit to work control the activity at the point of execution, which is where maintenance-induced outages are prevented. Reporting frequency, capital planning and retrospective service review operate on different timescales and do not control the work itself.

Question 18Data Centre Operations Management (operational processes, service level and KPI management)

A service review shows that mean time to restore has improved but customer satisfaction has fallen. Which explanation is most plausible?

  1. The metric excludes communication quality and frequency of updates during incidents Correct answer.
  2. Customers no longer care about restoration time
  3. Satisfaction surveys are always unreliable
  4. The restoration metric must be calculated incorrectly

Explanation

Restoration time captures only one dimension of the customer experience; poor communication, unclear status updates or repeated short outages can reduce satisfaction even when average restoration improves. Assuming calculation error, dismissing customer priorities or rejecting the survey data avoids the more likely operational explanation.

Question 19Data Centre Operations Management (operational processes, service level and KPI management)

Which set of KPIs best supports an operations team managing a critical facility on a daily basis?

  1. Number of marketing leads and conversion rate
  2. Employee satisfaction score and average tenure
  3. Real-time alarm status, capacity headroom, open critical work orders and incident status Correct answer.
  4. Annual revenue per rack and customer churn rate

Explanation

Daily operational control depends on current condition and workload indicators such as alarm status, capacity headroom, open critical work orders and live incident status. Revenue, marketing and workforce measures inform other management layers but do not support daily facility control.

Question 20Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator is preparing for a customer audit of its service level reporting. Which practice most strengthens confidence in the reported figures?

  1. Reporting figures prepared by the sales team from memory
  2. Restricting the audit to a single favourable month
  3. Presenting summary figures without supporting detail to protect commercial confidentiality
  4. Allowing the customer to recalculate figures from raw monitoring exports with a documented methodology Correct answer.

Explanation

Auditability depends on a documented methodology and access to the underlying raw data so the figures can be independently recalculated. Summary-only reporting, informal preparation and selective periods all weaken confidence in the reported performance.

Question 21Data Centre Operations Management (operational processes, service level and KPI management)

An operator has a target of 99.999 percent availability for the IT load. Which operational implication follows most directly from this target?

  1. The facility can rely on a single utility feed with no backup
  2. Concurrent maintainability and tightly controlled change windows become essential to avoid planned downtime counting against the target Correct answer.
  3. Planned maintenance can be performed at any time without affecting the target
  4. Reporting frequency can be reduced because outages will be rare

Explanation

At five-nines availability the annual downtime budget is only a few minutes, so planned work must be performed without interrupting the load, which requires concurrent maintainability and disciplined change windows. Assuming free maintenance windows, single-path supply or reduced reporting all conflict with the target.

Question 22Data Centre Operations Management (operational processes, service level and KPI management)

A data centre experienced four incidents in a year: two caused by human error during maintenance, one by a UPS component failure and one by a cooling pump failure. Which conclusion is best supported by this data?

  1. Human error during maintenance is the largest single cause category and warrants process controls Correct answer.
  2. Cooling failures are the dominant risk to availability
  3. The facility should stop all maintenance activity
  4. The UPS is the least reliable system in the facility

Explanation

Two of the four incidents share a maintenance human-error cause, making it the largest single category and pointing to process controls such as permits, procedures and supervision. Single equipment failures cannot be ranked as dominant from one event each, and halting maintenance would create greater risk.

Question 23Data Centre Operations Management (operational processes, service level and KPI management)

A service level agreement offers credits when monthly availability falls below target. The operations team proposes to measure availability excluding any incident lasting less than fifteen minutes. What is the main governance concern?

  1. The exclusion changes the agreed measurement basis and could allow repeated short outages to escape the commitment Correct answer.
  2. Credits should never be offered for availability shortfalls
  3. Short incidents are too difficult to record accurately
  4. Fifteen minutes is longer than any realistic outage

Explanation

Redefining the measurement basis after the agreement is set alters the commitment and can let frequent short outages accumulate real customer impact without triggering credits. The concern is governance and fairness of the metric, not recording difficulty, credit policy in principle or outage duration assumptions.

Question 24Data Centre Operations Management (operational processes, service level and KPI management)

An operator's capacity KPI shows 30 percent headroom on power but 5 percent on cooling. How should operations management interpret this?

  1. Headroom figures should be ignored until a customer requests additional load
  2. Power is the binding constraint and should drive sales commitments
  3. The facility is balanced because the average headroom is acceptable
  4. Cooling is the binding constraint and should govern new commitments and capacity planning Correct answer.

Explanation

Capacity is limited by the most constrained resource, so cooling at five percent headroom governs how much additional load can be accepted regardless of spare power. Averaging the figures, treating power as binding or deferring the analysis would risk overcommitting the facility.

Question 25Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator is deciding whether to report a near-miss in which a maintenance procedure nearly caused a UPS transfer to bypass. Which approach best reflects mature operational governance?

  1. Record and investigate the near-miss, share the learning and adjust the procedure Correct answer.
  2. Record it only if a customer later asks about it
  3. Discipline the technician and take no further action
  4. Omit it from reporting because no service impact occurred

Explanation

Near-misses are leading indicators of risk, and recording, investigating and sharing the learning prevents recurrence before an actual outage occurs. Omitting them, reporting on request or focusing only on individual blame forgoes the opportunity to correct the underlying procedure.

Question 26Data Centre Operations Management (operational processes, service level and KPI management)

In data centre service level management, what distinguishes an SLA from an internal operational level agreement (OLA)?

  1. The SLA is a contract with the customer carrying remedies, while the OLA binds internal teams that support delivery of that SLA. Correct answer.
  2. The SLA is optional documentation, while the OLA is the only document enforceable in a customer dispute.
  3. The SLA defines technical setpoints such as cold aisle temperature, while the OLA defines commercial credit amounts.
  4. The SLA is reviewed annually by facilities engineering, while the OLA is reviewed monthly by the customer.

Explanation

An SLA is the externally facing commitment to the customer, including measurable targets and consequences such as service credits. OLAs are internal agreements between supporting teams that allocate the responsibilities needed to meet the SLA. Confusing the two usually results in targets nobody has accepted operationally.

Question 27Data Centre Operations Management (operational processes, service level and KPI management)

An operations team wants a KPI that reveals whether the maintenance programme is being executed as scheduled rather than merely planned. Which measure is most appropriate?

  1. Ratio of planned maintenance hours to total staffed hours
  2. Number of maintenance work orders raised per technician per month
  3. Percentage of scheduled maintenance work orders completed within their planned window Correct answer.
  4. Total maintenance spend compared with the annual budget

Explanation

Schedule compliance measures completed-versus-planned work within the agreed window, which directly shows whether the programme is executed on time. Volume, labour ratio and spend describe effort or cost but say nothing about timeliness of execution. Schedule compliance is the standard reliability indicator for maintenance delivery.

Question 28Data Centre Operations Management (operational processes, service level and KPI management)

A colocation provider must report how quickly customer-requested cage installations are delivered. Which measurement approach is most defensible in the monthly service report?

  1. Report the average elapsed time from request to handover across all requests received that month
  2. Report the number of installations completed, without reference to the committed lead time
  3. Report the percentage of requests handed over within the committed lead time, with the sample size stated Correct answer.
  4. Report only the fastest quartile of installations to demonstrate best performance

Explanation

A percentage-within-commitment measure ties reporting directly to the agreed lead time and states the population, so the customer can judge reliability rather than a single average. Averages hide tail performance, and selective reporting of best cases undermines trust. Stating sample size prevents misleading conclusions from small volumes.

Question 29Data Centre Operations Management (operational processes, service level and KPI management)

Which statement best describes the role of a service catalogue in data centre operations management?

  1. It specifies the electrical topology of each data hall for capacity planning
  2. It defines the services operations delivers, with owners, descriptions and associated service levels Correct answer.
  3. It records the incident history used to calculate availability each month
  4. It lists the physical assets in the white space and their maintenance dates

Explanation

The service catalogue is the authoritative list of services provided, each with an owner, description and the service levels that apply. It is the reference point from which SLAs, OLAs and reporting are derived. Asset registers, incident logs and single-line diagrams are separate operational records.

Question 30Data Centre Operations Management (operational processes, service level and KPI management)

A monthly service review shows availability met target, but the number of repeat incidents from the same root cause rose sharply. What does this combination most likely indicate about the KPI set?

  1. The operations team should stop reporting availability until repeat incidents reach zero
  2. The KPI set is output-focused and lacks a measure of problem management effectiveness Correct answer.
  3. Availability is being measured incorrectly, because a met target cannot coexist with rising repeat incidents
  4. Repeat incidents are a customer-experience issue and should be excluded from operational reporting

Explanation

Availability can remain within target while the same failures recur, because each event is short enough not to breach the monthly figure. That pattern signals a gap in problem management, which requires its own KPI such as repeat-incident rate or problem closure rate. Adding that measure keeps the review balanced between outcome and learning.

Question 31Data Centre Operations Management (operational processes, service level and KPI management)

An operations manager is defining severity levels for incidents. Which basis produces the most consistent classification across shifts?

  1. The time of day the incident is reported and the number of staff on shift
  2. The seniority of the engineer who first takes the call
  3. The assessed business impact and urgency against written criteria, applied by the incident manager Correct answer.
  4. The number of alarms raised by the building management system

Explanation

Severity should reflect impact and urgency assessed against documented criteria, so different shifts reach the same classification for the same situation. Alarm counts, staffing and caller seniority are not measures of business impact. Written criteria also make escalation decisions auditable.

Question 32Data Centre Operations Management (operational processes, service level and KPI management)

A customer SLA promises 99.9 percent monthly availability of the critical power path. Operations wants an internal target that gives early warning before the SLA is breached. Which internal target is most sensible?

  1. Set the internal target more stringent than the SLA to create a buffer before customer impact Correct answer.
  2. Set the internal target lower than the SLA because internal measures are only indicative
  3. Remove the internal target and rely on the monthly customer calculation
  4. Set the internal target equal to the SLA so performance is reported honestly

Explanation

Internal targets are normally tighter than the customer commitment so that deterioration triggers action before the SLA is breached. Setting them equal leaves no warning margin, and setting them lower removes any protective intent. The internal measure should also be calculated on a comparable basis to the SLA.

Question 33Data Centre Operations Management (operational processes, service level and KPI management)

During a service review, a customer asks how planned maintenance downtime is treated in the availability KPI. Which treatment is most defensible?

  1. Include planned maintenance in the headline figure and disclose the split between planned and unplanned
  2. Report planned maintenance separately and state in the SLA whether it counts against the target Correct answer.
  3. Report availability only for months in which no planned maintenance occurred
  4. Exclude all maintenance windows from the calculation and report only unplanned downtime

Explanation

The correct treatment is whatever the SLA defines, and it must be disclosed so the figure can be interpreted. Reporting planned and unplanned downtime separately preserves transparency while keeping the headline figure consistent with the contract. Excluding or omitting months makes the KPI unverifiable.

Question 34Data Centre Operations Management (operational processes, service level and KPI management)

An operator wants to report capacity headroom in a way that supports decisions rather than simply describing the plant. Which KPI formulation best achieves this?

  1. Total installed electrical capacity in kilowatts for the whole site
  2. Nameplate rating of the cooling units serving the white space
  3. Percentage of usable capacity consumed and the months until a defined threshold is reached at current growth Correct answer.
  4. Number of racks currently occupied in each data hall

Explanation

A decision-supporting capacity KPI combines current utilisation with a projection of when a threshold will be crossed, so management can time build or procurement. Static installed figures and occupancy counts describe state but not urgency. Projected runway converts the measure into an actionable signal.

Question 35Data Centre Operations Management (operational processes, service level and KPI management)

A vendor performs maintenance under a service contract and causes an outage. Which contract and process control most directly reduces recurrence?

  1. Requiring a method statement and permit approval before work, with post-work verification recorded Correct answer.
  2. Increasing the penalty amount in the contract after the event
  3. Extending the vendor's contract term to encourage long-term accountability
  4. Adding the vendor to the monthly service review distribution list

Explanation

Pre-work method statements and permit control prevent unsafe work, and recorded post-work verification confirms the plant is restored correctly. Penalties act after the loss and do not by themselves change work practice. Communication and contract duration do not control the technical execution of the task.

Question 36Data Centre Operations Management (operational processes, service level and KPI management)

An operations team proposes to measure mean time between failures (MTBF) for the UPS fleet. What limitation must management understand when interpreting this KPI?

  1. MTBF is identical to mean time to restore and therefore adds no new information
  2. MTBF is an average that can mask a small number of units failing repeatedly and it rises when failures are simply not reported Correct answer.
  3. MTBF can only be calculated for mechanical equipment, not for electrical systems
  4. MTBF requires continuous customer surveys to be valid

Explanation

MTBF is an average across a population, so a few unreliable units can be hidden by many reliable ones. It also improves artificially if near-misses or short failures go unreported. Reporting failure counts and per-unit history alongside MTBF avoids these distortions.

Question 37Data Centre Operations Management (operational processes, service level and KPI management)

A data centre operator is introducing a daily operational review meeting. Which agenda design best supports control of a critical facility?

  1. Review the annual budget position and capital project milestones
  2. Review customer satisfaction scores and marketing pipeline
  3. Review yesterday's incidents and near-misses, open permits and change activity, current plant status and today's at-risk activities Correct answer.
  4. Review staff training records and annual leave balances

Explanation

A daily operations review should focus on the live risk picture: what failed, what nearly failed, what work is authorised today and which plant is degraded. That allows the team to set priorities before work begins. Financial, commercial and HR topics belong in separate governance forums.

Question 38Data Centre Operations Management (operational processes, service level and KPI management)

An operator claims 99.999 percent availability for the IT load but calculates it from utility supply availability only. Why is this claim analytically weak?

  1. Availability should always be expressed as downtime hours rather than a percentage
  2. The calculation omits the UPS, switchgear, cooling and distribution paths that actually determine IT load availability Correct answer.
  3. Utility supply is the only element that can be measured reliably, so no other basis is possible
  4. Five nines is not an accepted availability figure in the data centre industry

Explanation

IT load availability depends on the whole critical path, including UPS, distribution and cooling, not just the incoming utility. Measuring only utility supply can produce a figure far better than the actual customer experience. The KPI must be defined at the point of delivery and include all elements in the agreed boundary.

Question 39Data Centre Operations Management (operational processes, service level and KPI management)

A provider offers service credits for missing a monthly availability target. Operations proposes to maximise reported availability by reclassifying certain events as customer-caused. What governance control best protects the integrity of the reported figure?

  1. Allow the operations manager to decide classifications, since they hold the technical knowledge
  2. Suspend credits whenever the cause is disputed until an annual review
  3. Require event classification against pre-agreed definitions with evidence, reviewed and signed off jointly with the customer Correct answer.
  4. Report the higher of the operations and customer calculations each month

Explanation

Classification rules agreed in advance, applied with evidence and jointly reviewed, prevent disputes and self-interested reclassification. Unilateral decisions by the party being measured create a conflict of interest. Choosing the higher figure or deferring disputes undermines the credibility of the whole reporting regime.

Question 40Data Centre Operations Management (operational processes, service level and KPI management)

An operations team reports a rising number of near-misses alongside a falling incident rate. How should management interpret and act on this pattern?

  1. Ignore near-miss data until an actual incident confirms the trend
  2. Recognise that increased near-miss reporting can indicate a healthier reporting culture, and analyse the reports for common precursors Correct answer.
  3. Treat the near-miss rise as evidence that reporting culture is weak and discourage low-severity reports
  4. Conclude that safety performance is deteriorating because near-misses are early-stage incidents

Explanation

Near-miss counts rise when reporting barriers fall, so the trend must be read alongside incident rates and reporting participation. The value lies in analysing precursors and closing them before harm occurs. Discouraging reports or waiting for a real incident wastes the earliest warning available.

Question 41Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

In planned preventive maintenance for a data centre's CRAC units, which document defines the task list, frequency, and acceptance criteria that the maintenance contractor must follow?

  1. The facility's incident register for the past year
  2. The staff shift roster for the operations team
  3. The vendor's marketing brochure for the CRAC model
  4. The maintenance contract scope and its referenced task schedules Correct answer.

Explanation

The contract scope together with the referenced task schedules is the controlling document; it sets what is done, how often, and how completion is judged. A brochure is promotional, an incident register is historical evidence, and a roster is staffing, so none of them defines contractual maintenance obligations.

Question 42Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

A data centre operator signs a contract where the contractor is paid a fixed monthly sum regardless of how many callouts occur. Which contract type is this?

  1. Fixed-price lump-sum contract Correct answer.
  2. Time-and-materials contract
  3. Performance-based bonus contract
  4. Cost-plus contract

Explanation

A fixed monthly payment independent of the number of callouts is a lump-sum arrangement, placing volume risk on the contractor. Time-and-materials and cost-plus both charge according to work performed or cost incurred, so the operator would carry that risk.

Question 43Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

Which maintenance strategy is most appropriate for a UPS battery string whose failure mode is gradual capacity degradation with a measurable trend?

  1. Deferred maintenance until the next facility refurbishment
  2. Calendar-based replacement of the whole string every month
  3. Run-to-failure maintenance with no intervention until the string fails
  4. Condition-based maintenance driven by periodic impedance and capacity tests Correct answer.

Explanation

Battery degradation is progressive and observable, so periodic impedance and capacity testing lets the operator replace the string before it fails on load. Run-to-failure risks an unprotected load, monthly replacement is economically absurd, and deferral ignores a known degradation mechanism.

Question 44Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

What is the primary purpose of a permit-to-work system for maintenance activities in a live data centre?

  1. To schedule the annual budget review for the facilities team
  2. To formally authorise specific work after hazards and isolations are verified Correct answer.
  3. To track spare-parts stock levels in the warehouse
  4. To record the contractor's invoicing details for the maintenance visit

Explanation

A permit-to-work is a formal control that authorises defined work only after hazards are assessed and isolations or safeguards are confirmed. It is a safety and change-control instrument, not a commercial, financial, or inventory record.

Question 45Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

During a generator load test, the maintenance technician observes that the fuel transfer pump runs continuously. What is the most appropriate first corrective action under a corrective maintenance work order?

  1. Investigate the transfer pump control circuit and float or level switch operation Correct answer.
  2. Immediately replace the entire generator set
  3. Close the work order as no fault is present
  4. Increase the test load to force the pump to stop

Explanation

Continuous running points to a control or level-sensing fault, so diagnosis should start at the pump control circuit and level devices. Replacing the generator set is grossly disproportionate, adding load does not address the cause, and closing the order would leave a latent fuel-system fault.

Question 46Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

Which key performance indicator best measures whether a maintenance contractor is completing the agreed preventive maintenance workload?

  1. PM schedule compliance percentage Correct answer.
  2. Number of contractor staff on site
  3. Mean time between failures of the cooling plant
  4. Total annual maintenance spend

Explanation

Schedule compliance compares completed PM tasks against those due in the period, directly measuring delivery of the agreed workload. MTBF reflects reliability outcomes, headcount is a resourcing input, and spend is a cost measure, none of which shows whether planned tasks were done.

Question 47Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

A maintenance contract requires the contractor to respond on site within four hours for critical cooling failures. Which term describes this obligation?

  1. Service level agreement response time Correct answer.
  2. Planned maintenance frequency
  3. Mean time to repair
  4. Spare parts holding level

Explanation

A contractual commitment to attend within a stated time is a service level agreement response time, and it is measured from fault notification to arrival. MTTR measures repair duration, PM frequency sets task intervals, and spares holding is an inventory commitment.

Question 48Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

Which record is the most reliable source for demonstrating that a specific maintenance task was actually performed on a given date?

  1. A signed and time-stamped work order with technician sign-off Correct answer.
  2. A verbal confirmation from the contractor's account manager
  3. The contractor's annual summary presentation
  4. A purchase order raised for the maintenance visit

Explanation

A signed, time-stamped work order ties a named task to a date and a responsible technician, forming contemporaneous evidence. Verbal confirmation is not auditable, an annual summary aggregates without per-task proof, and a purchase order only shows commercial intent.

Question 49Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

A data centre plans to move from purely calendar-based generator servicing to a regime that also uses run-hours and oil analysis results. Which maintenance approach does this describe?

  1. Pure run-to-failure maintenance
  2. Deferred corrective maintenance
  3. Reactive breakdown maintenance
  4. A hybrid of time-based and condition-based maintenance Correct answer.

Explanation

Combining fixed calendar intervals with usage-based and condition-monitoring triggers is a hybrid time-based and condition-based approach. The other options all describe waiting for failure or postponing work, which is the opposite of this planned regime.

Question 50Facilities Management and Maintenance (planned and corrective maintenance, maintenance contracts)

Under a maintenance contract, the contractor is required to hold critical spares on site. Which item is the strongest candidate for on-site holding in a data centre cooling system?

  1. A spare chiller of identical capacity
  2. A full set of building management system servers
  3. A chilled water pump mechanical seal kit matched to the installed pumps Correct answer.
  4. A complete replacement cooling tower

Explanation

On-site spares should target items with high failure probability, long lead times, and low unit cost, such as a matched mechanical seal kit. Whole chillers, cooling towers, and BMS servers are capital-intensive and normally covered by redundancy or vendor support rather than on-site stock.