CDCS EXIN EPI Certified Data Centre Specialist Questions and Answers

When calculating Power Usage Effectiveness (PUE) in a data center that uses chilled water from an external source, like from a building owner, a weight factor for district chilled water must be applied. This is because PUE calculations aim to measure the energy efficiency of the data center’s own operations, and external utilities like district chilled water aren’t directly powered by the data center. A weight factor of 0.4 is typically used to account for the energy consumed to produce and deliver the chilled water, reflecting the indirect impact on the data center’s total energy consumption.

Detailed Explanation:

PUE is calculated as the ratio of the total facility energy to the IT equipment energy. If the cooling is provided by an external chilled water source, it’s necessary to adjust the calculations to accurately reflect the energy impact. By incorporating the 0.4 weight factor, data centers can calculate a more accurate PUE, aligning with standard methods and industry best practices.

EPI Data Center Specialist References:

EPI training on PUE highlights the importance of adjusting for external energy sources, such as district cooling, in the calculations. This ensures that PUE values remain accurate and comparable across different data centers, even when external utilities are used.

The formula used to calculate the gas content differs significantly between halocarbon and inert-based fire suppression systems. Halocarbon systems function by absorbing heat, while inert systems work by reducing oxygen levels. Due to these differences in fire suppression mechanisms, distinct formulas are applied, factoring in the specific properties of each gas type and the required concentration levels.

Detailed Explanation:

Halocarbon systems like FM200 require a formula that accounts for the concentration needed for cooling, while inert gases like nitrogen or argon need a formula that calculates the volume based on oxygen displacement. As the design concentration and characteristics of these gases differ, it’s essential to use the correct formula specific to the gas type.

EPI Data Center Specialist References:

EPI recommends consulting the specific design requirements and formulas provided by each gas manufacturer when switching fire suppression systems to ensure the correct amount of gas is deployed for effective fire suppression.

ANSI/TIA-942 specifies minimum floor live load capacities based on Rated levels:

Rated-1/2: ~7.2–8.4 kPa

Rated-3: 12 kPa

Rated-4: 15 kPa

These values ensure raised floors can support racks, cabling trays, cooling units, and maintenance loads without structural compromise. For Rated-3, concurrent maintainability requires higher floor robustness to handle additional infrastructure.

Therefore, the correct requirement is 12 kPa.

The diagram shows three UPS modules, each 100 kW, connected in parallel to support a 100 kW IT load. That means:

One module (100 kW) can support the load (N).

Two additional modules are installed as redundancy.

This equals N+2 redundancy.

2+N+1 and 2(N+1) imply dual active paths not shown.

N+N+N is not an industry term.

Thus, the correct redundancy level is N+2.

The most effective EMF mitigation is to install shielding as close as possible to the source of radiation. By blocking or redirecting magnetic flux at the origin (the transformer room walls), the overall field propagation into adjacent areas is minimized. If shielding were placed at the computer room, the field would already have spread over the intervening space, requiring more material and higher cost.

Standards such as IEEE Std 299 (EMC Shielding Effectiveness) and IEC 61000 emphasize source-based mitigation. Additionally, ANSI/TIA-942 requires EMF shielding where magnetic flux exceeds recommended ICT thresholds (generally <5 mG for sensitive tape/disk storage).

Although 100 mG is often tolerated by modern equipment, legacy magnetic storage can be affected, so shielding is still prudent. Hence, the correct location is at the transformer room wall.

Earth Leakage Protection is primarily installed to protect human lives by detecting and disconnecting power when a fault current flows to the ground. This type of protection is essential to avoid electrical shock hazards that could occur when insulation fails, or equipment is improperly grounded.

Detailed Explanation:

Earth leakage currents can occur due to insulation faults or accidental contact with live parts. Earth Leakage Protection systems, such as Residual Current Devices (RCDs), quickly detect these faults and disconnect the circuit to prevent harm to personnel. This is especially crucial in environments like data centers where high-powered equipment is continuously running and any electrical fault can pose significant safety risks.

EPI Data Center Specialist References:

EPI emphasizes that human safety is paramount in data center operations. Proper grounding and leakage protection are fundamental safety measures, and EPI guidelines align with this focus, underscoring the importance of protecting personnel from electrical hazards through appropriate safety systems.

In seismic-prone areas, cables should be tightly secured to racks and cable trays to minimize movement during seismic activity. Properly securing the cables prevents them from being damaged due to excessive motion, which could lead to outages or damage to connected equipment.

Detailed Explanation:

Loose cables can be vulnerable to shaking or sudden jolts during an earthquake, increasing the risk of disconnection or damage. By tightly organizing and securing cables, you ensure they remain in place, even during significant movement, thereby maintaining connection integrity and reducing the risk of physical damage.

EPI Data Center Specialist References:

EPI training includes considerations for data centers in seismic zones, advising that cables be firmly anchored to support structures to reduce movement and mitigate potential damage during seismic events.

The attenuation factor for shielding material is typically calculated using the formula A = 20 log (R / M). This equation provides the attenuation in decibels (dB), where R represents the measured electromagnetic field strength, and M is the maximum acceptable level. The logarithmic scale helps quantify how much the shielding reduces EMF levels relative to the maximum allowable value.

Detailed Explanation:

This formula calculates attenuation by comparing the measured value with the acceptable threshold, with the result expressed in decibels. A higher attenuation indicates more effective shielding material, essential for environments requiring robust EMF management.

EPI Data Center Specialist References:

EPI standards include the use of logarithmic formulas to evaluate attenuation levels, ensuring that shielding materials provide adequate reduction in EMF to protect sensitive equipment within data centers.

The server’s repeated overheating despite adequate intake and exhaust temperatures suggests that dust buildup inside the server may be impeding heat transfer. Dust accumulation can obstruct airflow within the server, insulate components, and disrupt the convection-based cooling systems that regulate internal temperatures, leading to overheating and potential hardware failures.

Detailed Explanation:

While the intake and exhaust temperatures appear within acceptable ranges, internal dust can reduce airflow and impede cooling efficiency, causing internal components to overheat despite seemingly normal ambient conditions. Regular cleaning and maintenance are critical for preventing dust-related issues, especially in high-powered equipment like a 5kW server.

EPI Data Center Specialist References:

EPI emphasizes regular maintenance to prevent dust buildup in data center equipment. Dust can significantly impact cooling efficiency and lead to overheating, which underlines the importance of routine cleaning for optimal server performance.

MPO (Multi-fiber Push On) pre-terminated fiber offers the advantage of quick installation and supports high-speed upgrades up to 40 Gbit/s and 100 Gbit/s. This technology simplifies connections by allowing multiple fibers to be connected in a single plug-and-play module, reducing installation time and making future expansions or upgrades to higher speeds easier.

Detailed Explanation:

MPO connectors consolidate multiple fiber connections into a single interface, which simplifies cable management and reduces installation complexity. This setup is particularly beneficial in data centers where high-speed networks are essential, and where rapid deployment is necessary. MPO pre-terminated fiber also supports the use of OM4 fiber, making it compatible with existing high-performance cabling infrastructure.

EPI Data Center Specialist References:

EPI training highlights that pre-terminated fiber solutions, like MPO, can greatly enhance efficiency in data centers. By reducing installation time and providing scalability for higher transmission speeds, they align with best practices for maintaining flexible and future-ready network infrastructure.

The electrical design shown in the diagrams represents a TIA-942 Rating-4 configuration. This design includes full redundancy and fault tolerance, as demonstrated by the dual power distribution paths from the utility supply to the critical loads. Each power distribution path is equipped with its own UPS, ensuring that the ICT equipment and mechanical equipment have uninterrupted power in case of any single point of failure.

Detailed Explanation:

A Rating-4 data center requires two independent power paths that are fully redundant and capable of supporting the load independently. In the diagrams:

There are dual feeds from the utility supply, each going through separate transfer switches and power distribution paths.

Both paths have backup sources (+1) and serve critical components through separate UPS systems, providing a completely redundant setup.

The design also includes redundant paths to the mechanical equipment and ICT equipment, which further indicates the fault-tolerant characteristics of a Rating-4 infrastructure.

This setup allows for concurrent maintainability and ensures that no single failure in power distribution or UPS can impact the data center’s operation, which is characteristic of the highest Tier/Rated-4 classification.

EPI Data Center Specialist References:

EPI guidelines confirm that TIA-942 Rating-4 requires full redundancy and fault tolerance for electrical infrastructure, ensuring continuous operation even during maintenance or failure events. This design meets all those requirements, thus aligning with Rating-4 standards.

LCC (Life Cycle Cost) is the metric that accounts for capital cost, operational energy, maintenance, and disposal/replacement. This holistic view is critical when evaluating cooling technologies like CRACs, CRAHs, chilled water, or free cooling systems.

HER (Heat Exchange Ratio) and COP (Coefficient of Performance) measure operational efficiency but not total lifecycle costs.

PUE measures overall data center efficiency but not specifically cooling CAPEX/OPEX.

Thus, LCC is the comprehensive metric that covers installation, operation, and disposal.

Inert gas systems (e.g., Inergen, Argonite, Nitrogen) extinguish fire by reducing oxygen concentration, which typically requires reducing oxygen levels to ~12–15%. This means a very large volume of gas must be discharged into the room (up to 40–50% of the protected volume). Because inert gases have a low extinguishing effectiveness by weight, more total gas is required.

Halocarbon agents (FM-200, Novec 1230) extinguish fire chemically by interrupting the combustion chain reaction. They require only a small percentage (6–9%) concentration in the room volume. As a result, the storage space for cylinders is much smaller compared to inert gas systems.

Therefore, inert gas systems generally require a larger gas volume to achieve extinguishing concentrations.

According to ANSI/TIA-942 standards, the overall data center rating is determined by the lowest rating among all evaluated categories. Therefore, if a data center is rated 2 in architectural, despite being rated higher in mechanical and electrical, the overall rating is Rated-2. This approach ensures that all aspects meet a minimum standard and prevents a higher rating if any critical area does not comply.

Detailed Explanation:

ANSI/TIA-942 evaluates data centers across several areas, including mechanical, electrical, architectural, and telecommunications. The overall rating reflects the lowest rated category, ensuring that no aspect of the data center’s design or operation falls below the specified level. Thus, in this case, the architectural rating of 2 dictates the final rating, ensuring a comprehensive and balanced assessment of reliability and resilience across all aspects.

EPI Data Center Specialist References:

EPI Data Center Specialist training aligns with ANSI/TIA-942, stating that the final rating must reflect the lowest score to ensure comprehensive reliability across all critical infrastructure categories. This avoids overstating the data center's resilience and ensures uniform standards across areas.

“Overcurrent” for protective devices is the condition above the device’s rated/trip value that causes operation (tripping/melting).

A = inrush tolerance, not the definition of overcurrent.

B = prospective short-circuit current (Ik).

ASHRAE TC 9.9 recommends using dust deposition coupons (also called reactivity coupons) to measure particulate contamination over time. These are small, standardized metallic plates exposed to the air that collect dust and contaminants. After a set period, they are analyzed for chemical corrosion rates and particulate buildup.

Option A (handheld analyzers) gives spot counts of airborne particles but does not capture long-term corrosive deposition risk.

Option C only samples underfloor plenum dust, not actual IT environment contamination.

Option D (opening equipment) is invasive and risks voiding warranties.

Coupons provide standardized, passive, and reliable monitoring of contamination over months, making them the preferred method.

Raised floors must not be installed flush against perimeter walls. Buildings expand, contract, and shift due to thermal expansion (hygrothermal movement), structural settlement, and seismic activity. If the raised floor is rigidly connected to the walls, this movement transfers directly into the raised floor grid, causing buckling, cracking, or tile misalignment.

Industry standards (e.g., CISCA – Ceilings & Interior Systems Construction Association guidelines) require leaving a small perimeter gap (typically 1–2 cm) between raised floor tiles and walls. This gap is then sealed with flexible firestops or brush gaskets to control air leakage while preserving movement tolerance.

Therefore, the correct practice is to leave a separation space, making answer C correct. Options A and D restrict this requirement incorrectly, while B is factually wrong (sealing against the perimeter is not allowed).

From an efficiency standpoint, Unit B with a Sensible Heat Ratio (SHR) of 0.9 is preferable. A higher SHR indicates that a greater proportion of the air conditioner's capacity is dedicated to sensible cooling (temperature reduction) rather than latent cooling (moisture removal). In data centers, sensible cooling is more critical since IT equipment primarily generates heat without adding significant moisture.

Detailed Explanation:

An SHR of 0.9 means that 90% of the cooling capacity is used for sensible cooling, which is more efficient for environments like data centers where humidity control is typically less of a concern. Opting for an air conditioner with a higher SHR ensures that most of the cooling energy is focused on temperature reduction, making Unit B more efficient in this scenario.

EPI Data Center Specialist References:

EPI data center best practices recommend choosing cooling units with higher SHR values in data centers, as they better match the cooling needs of IT equipment. High SHR units improve cooling efficiency by concentrating on sensible heat removal, which is vital for maintaining the optimal thermal environment.

When installing a raised floor, the best starting point is typically from the farthest corner away from the entrance, in this case, Point B. This method allows for the floor installation to progress towards the entrance, ensuring that workers can avoid walking over the newly installed tiles, thus minimizing the risk of damage and maintaining cleanliness during installation.

Detailed Explanation:

Starting at the farthest corner (Point B) and working towards the entrance (Point A) is a standard practice in raised floor installation. This approach ensures that installers do not walk over freshly laid tiles, which could lead to shifting or misalignment. Additionally, it allows for more controlled placement and alignment, as well as convenient egress when installation is complete.

EPI Data Center Specialist References:

EPI guidelines for raised floor installation emphasize beginning at the farthest point from the entrance and working back towards it, which aligns with industry best practices for efficiency and quality assurance. This method reduces potential damage and supports precise alignment across the floor area.

The cooling capacity of a raised floor tile is primarily influenced by the percentage of surface opening, the obstruction caused by the supporting construction, the pressure under the raised floor, and the damper construction. These factors dictate how much airflow can pass through the tile and how effectively cool air is distributed to the equipment in the data center.

Detailed Explanation:

The percentage of surface opening on a floor tile affects how much air can flow through, with larger openings allowing more airflow. Supporting structures beneath the floor can obstruct airflow, reducing cooling efficiency. Pressure under the raised floor impacts the velocity and volume of air that moves through the tile. Additionally, if dampers are installed, they control the airflow rate, which can be adjusted to meet specific cooling needs for the area.

EPI Data Center Specialist References:

EPI guidance on airflow management under raised floors emphasizes these factors as critical for effective cooling, especially in high-density areas. Ensuring unobstructed and adequate airflow helps maintain consistent cooling across equipment.

For a cooling audit in a data center, it is essential to measure temperature and humidity where air enters the servers to accurately assess cooling performance. In this case, since all servers have a front-to-rear (F-R) airflow design, measuring at the front/intake of the server will provide a precise understanding of the cooling conditions that the equipment is experiencing.

Detailed Explanation:

Servers with a front-to-rear airflow design draw in cool air from the cold aisle at the front, which is then exhausted into the hot aisle at the rear. By measuring temperature and humidity 50 mm/2 inches from the front intake, you gather data on the air conditions right before it enters the servers, providing an accurate representation of the cooling environment as it directly impacts the equipment.

Measuring in the cold aisle at the front intake ensures that the readings reflect the actual conditions of the incoming air that the servers depend on for effective cooling. This approach is consistent with best practices for maintaining thermal conditions in a data center, as it helps confirm that the cooling systems are delivering air within the required temperature and humidity specifications.

EPI Data Center Specialist References:

According to the EPI Data Center Specialist curriculum, the optimal placement for temperature and humidity sensors is at the intake of the equipment in the cold aisle, as it directly correlates to the environmental conditions affecting the servers. This positioning allows for a more effective audit of cooling performance, which is critical for maintaining the reliability and efficiency of the data center’s operations.

Forced-entry ratings use defined attack tools and durations (time-to-breach) executed by attacker profiles.

When the window area exceeds the maximum exposed area specified for fire-rated glass, it is necessary to split the window into sections that comply with the fire rating requirements. This means creating smaller sections that are each within the 3 sqm/32 sqft limit and using fire-rated frames to ensure that the entire assembly meets fire safety standards. This approach maintains the fire-rated integrity of the glass, while allowing for larger window areas.

Detailed Explanation:

Fire-rated glass is designed to contain fire and prevent it from spreading. If the window exceeds the maximum exposed area defined by the supplier, the integrity of the fire-rated glass could be compromised. By dividing the window into compliant sections with fire-rated frames, you ensure that each pane performs as intended in the event of a fire. Fire-rated frames help maintain the fire resistance across the entire assembly, making this option the best for safety and compliance.

EPI Data Center Specialist References:

EPI recommends adhering strictly to fire safety standards, especially when using materials like fire-rated glass. The guidelines emphasize that modifications should always respect the manufacturer’s specifications to ensure the system remains effective in containing and preventing the spread of fire.

Hydrogen sulfide (H₂S) is a corrosive gas that readily reacts with metals, especially copper and silver found in circuit boards, connectors, and power supply components. Even at concentrations below human detection thresholds, H₂S can cause sulfidation corrosion leading to intermittent connections, failure of solder joints, and increased failure rates of electronic equipment.

ASHRAE Technical Committee 9.9 has documented multiple failures in data centers due to corrosive gases (notably sulfur-bearing compounds). Therefore, high humidity combined with H₂S presence accelerates this risk.

H₂S does not affect static flooring properties (B) or fire suppression agent chemistry (D). Option A is clearly incorrect as the risk is well-documented.

An imbalance in load sharing between UPS systems connected in parallel can often result from unequal cable lengths to the common busbar. If the cabling from each UPS to the busbar varies significantly in length, it can lead to differences in impedance, resulting in uneven load distribution. Ensuring that cable lengths are consistent helps to balance the load sharing across the UPS systems.

Detailed Explanation:

Parallel UPS systems rely on uniform impedance to share loads evenly. Differences in cable lengths cause variations in resistance, leading to one or more UPS units carrying a disproportionate share of the load. Standardizing cable lengths ensures equal impedance, which promotes balanced load sharing and prevents one UPS from being overburdened, thus maintaining overall system reliability.

EPI Data Center Specialist References:

EPI guidelines recommend checking cable lengths when load imbalances occur in parallel UPS configurations. Ensuring equal lengths is a common method to resolve impedance issues that affect load distribution, which is critical for the stable operation of redundant power systems.

OM5 is optimized for Short Wavelength Division Multiplexing (SWDM) between 850–953 nm. This allows transmission of multiple wavelengths over a single fiber pair, reducing the number of fibers required for high-speed links.

OM4 already supports 100 GbE to 150 m, but OM5 with SWDM extends reach and reduces cabling bulk.

Option B is false because OM5 offers distinct SWDM benefits.

Option C is incorrect—OM5 is laser-optimized, not LED-based.

Option D is misleading; OM5 doesn’t extend 100 GbE to 500 m (that requires single-mode OS2 fiber).

Thus, the key advantage is SWDM support.

With infrequent/short outages, Standby (Emergency Standby Power, ESP) generators are appropriate; N+1 mitigates failure-to-start risk and supports concurrently maintainable designs.

Aerosol sprays release particulates and residues into the environment, which can contaminate sensitive ICT equipment. Such contamination accelerates corrosion, interferes with airflow, and increases particulate levels beyond ASHRAE recommended limits.

Proper cleaning should use HEPA-filtered vacuum systems or dry wipes, not aerosols. Even odor control aerosols are disallowed in critical rooms.

Therefore, aerosol cleaning solutions must never be used.

With Halocarbon fire suppression systems, as the temperature increases, the amount of gas required for effective suppression decreases. This is because Halocarbon agents are stored as a liquid and discharge as a gas, expanding more at higher temperatures. As a result, less agent is needed at higher room temperatures to achieve the desired concentration for fire suppression.

Detailed Explanation:

Halocarbons rely on specific volumetric concentrations to suppress fires. Higher temperatures cause the agent to expand more rapidly, effectively filling the protected area with less agent needed to reach the required concentration. This is in contrast to some other gases, where temperature changes might not have the same effect on discharge quantities.

EPI Data Center Specialist References:

EPI data center training on fire suppression indicates that understanding the physical properties of agents like Halocarbons is key for correct system sizing. As the temperature rises, the gas expands more readily, thus requiring adjustments in the amount needed for effective coverage.

Halocarbon agents such as FM-200 (HFC-227ea) and Novec-1230 (FK-5-1-12) are both defined under NFPA 2001 and ISO 14520 as clean agents, but their required design concentrations and physical properties differ. When calculating agent quantity, the minimum extinguishing concentration (MEC) and safety factor (S) must be taken into account. The S-factor is specific to each agent and reflects differences in molecular weight, density, and flame suppression chemistry.

For Novec-1230, the required design concentration is generally lower than for FM-200 (around 4.5–6% vs. 7–9%), but the calculation formula is the same except for substituting the correct S-factor. Therefore, you cannot reuse the exact formula parameters from FM-200; you must change the S-factor and apply Novec-1230’s physical constants.

This ensures compliance with NFPA 2001 Annex C, which provides correction formulas for room volume, temperature, and specific agent type. Using the wrong S-factor could result in underfilling or overfilling, compromising fire safety or increasing cost unnecessarily.

To determine the required capacity of the battery bank for the 15-minute runtime at full load, one must know the total power requirement that the battery bank must supply. The specifications provided include most of the necessary details, such as rated power, input voltage, battery voltage, and discharge voltage. However, one critical piece of information is missing: the UPS efficiency.

Detailed Explanation:

In a data center UPS system, the battery bank is designed to supply power for a set duration when there is an input power failure. The UPS efficiency affects the actual power the UPS can deliver to the load compared to the power it draws from the batteries. The efficiency factor is necessary to accurately calculate the required capacity of the battery bank since it determines how much input power is needed from the batteries to supply the load at full capacity. The formula typically used to determine battery capacity involves factoring in UPS efficiency, as it allows you to understand the losses within the UPS system.

If UPS efficiency is not considered, there would be an inaccurate estimation of the actual power needed from the batteries. For instance, if a UPS has 90% efficiency, only 90% of the power drawn from the batteries reaches the load. Without knowing this efficiency, it is not possible to calculate the battery bank size accurately, as you cannot accurately estimate the losses within the UPS itself.

EPI Data Center Specialist References:

According to EPI Data Center Specialist training, understanding the UPS efficiency is essential for battery sizing. Without it, the calculations could lead to either undersizing or oversizing the battery bank, which affects both reliability and cost-effectiveness of the UPS system. The EPI Data Center Specialist course emphasizes that battery sizing must account for all losses within the UPS system, with efficiency being a primary factor in these calculations.

The Top-of-Rack (ToR) cabling layout places an access switch directly inside each rack. Each server in that rack only requires a short patch cable to connect to the switch, and only one or two uplinks per rack connect to aggregation switches. This greatly reduces the number of long horizontal cables across the data hall.

In contrast, an End-of-Row (EoR) design centralizes switches at the row end, requiring long horizontal cables from each server to the row cabinet. This can lead to thousands of extra copper or fiber runs in large deployments.

ANSI/TIA-942 and Cisco Design Guides emphasize that ToR is the best solution for minimizing cable bulk, improving airflow, and reducing cost in hyperscale or dense rack environments.

Thus, if the explicit design goal is to minimize cable quantity, ToR design is superior.

In areas with frequent and extended power outages, continuous generators with at least an N+1 configuration are necessary to ensure consistent power availability. Continuous generators are designed for prolonged operation, making them suitable for scenarios where utility power is frequently unavailable, as in this case with outages exceeding 650 hours per year. An N+1 configuration ensures redundancy, which is critical for maintaining uptime in a high-availability data center.

Detailed Explanation:

Continuous generators provide reliable power over long durations, unlike standby generators, which are intended only for short-term use. The N+1 configuration ensures that there is always an additional generator available in case of failure, thus maintaining power supply even if one generator goes offline.

EPI Data Center Specialist References:

EPI best practices recommend continuous generators with redundancy for data centers located in areas with high power instability to maintain reliability and continuous operation.