CDCP Data Center Design Principles 2 — Questions and Answers
Question 1: The Uptime Institute's Tier IV data center classification requires what minimum percentage of concurrent maintainability?
- All paths and components are fully redundant with fault tolerance (2N or better) (Correct answer)
- N+1 redundancy on all mechanical and electrical systems
- 2N redundancy on power only; N+1 on cooling
- Dual power feeds with single cooling path
Correct answer: All paths and components are fully redundant with fault tolerance (2N or better)
Tier IV requires fully fault-tolerant systems, meaning all components and distribution paths are fully redundant (2N or better) so that any single equipment failure or planned maintenance does not interrupt the facility.
The Uptime Institute Tier Standard defines four tiers of data center reliability. Tier IV (Fault Tolerant) requires 2N or higher redundancy on all power and cooling components and distribution paths. Any single failure — planned or unplanned — must not impact IT load. Tier IV facilities target 99.9995% availability (26.3 minutes downtime/year). Tier III (Concurrently Maintainable) requires N+1 with multiple active paths. Understanding the differences between Tier I through Tier IV is a core CDCP exam competency.
Question 2: In data center design, what is the significance of a 'point of delivery' (POD) architecture?
- It modularizes the data center into self-contained, repeatable units that can be deployed incrementally as capacity grows (Correct answer)
- It defines the location where utility power enters the building
- It identifies the physical point where internet connectivity terminates inside the facility
- It sets the boundary between the data center operator's responsibility and the colocation customer's responsibility
Correct answer: It modularizes the data center into self-contained, repeatable units that can be deployed incrementally as capacity grows
A POD architecture divides the data center into modular, self-contained units with dedicated power, cooling, and networking, enabling incremental capacity expansion and faster deployment.
POD (Point of Delivery) architecture divides large data centers into repeatable, self-contained modules. Each POD includes a defined number of racks with its own dedicated power distribution (UPS, PDUs), cooling (CRAH units, CDUs), and network connectivity. This approach enables operators to build and commission one POD at a time rather than the entire facility, reducing upfront capital expenditure. It also simplifies capacity planning and allows for predictable, rapid expansion. Cloud and hyperscale operators pioneered this design approach.
Question 3: Which standard provides guidance specifically for data center telecommunications infrastructure design?
- TIA-942 (Correct answer)
- ASHRAE TC 9.9
- IEC 62443
- ISO/IEC 27001
Correct answer: TIA-942
TIA-942 (Telecommunications Infrastructure Standard for Data Centers) provides detailed guidance on cabling, layout, and redundancy for data center telecommunications infrastructure.
TIA-942 (now in its 2017 revision as TIA-942-B) defines the requirements for data center telecommunications cabling infrastructure. It covers the functional spaces of a data center (entrance room, main distribution area, horizontal distribution area, zone distribution area, and equipment distribution area), cabling topology, redundancy requirements, and alignment with the Uptime Institute's Tier ratings. CDCP candidates should know that TIA-942 Rated-3 and Rated-4 roughly align with Tier III and Tier IV concepts.
Question 4: What is 'white space' in data center terminology?
- The raised-floor area where IT equipment (servers, storage, networking) is located (Correct answer)
- Unused rack space available for future equipment installation
- The gap between hot and cold aisles in a containment system
- Network bandwidth capacity that has not been allocated to customers
Correct answer: The raised-floor area where IT equipment (servers, storage, networking) is located
White space refers to the primary data center floor area where IT equipment is installed, as opposed to 'gray space' (support infrastructure like UPS and generators) and office areas.
Data center real estate is divided into zones. White space is the raised-floor or flat-floor area where IT equipment resides — racks of servers, storage arrays, and network switches. Gray space houses mechanical and electrical support infrastructure: UPS systems, PDUs, generators, cooling equipment, and fire suppression. Office and administrative areas are separate. When operators discuss capacity, they measure white space in square feet/meters and kW of IT load. Colocation pricing is typically based on white space (per rack, per kW, or per square foot).
Question 5: When designing a data center, what is the primary purpose of the 'main distribution area' (MDA)?
- It is the core cabling hub where horizontal cabling from zone and equipment areas terminates, housing main cross-connects (Correct answer)
- It is the zone where power enters the building from the utility provider
- It is a dedicated space for network operations center (NOC) staff
- It is the area where backup generators are connected to the building switchgear
Correct answer: It is the core cabling hub where horizontal cabling from zone and equipment areas terminates, housing main cross-connects
The MDA is the cabling backbone of the data center — it houses the main cross-connect and is the central point from which horizontal cabling distributes to all other areas of the facility.
Per TIA-942, the data center cabling topology includes a hierarchical structure: Entrance Room (ER) → Main Distribution Area (MDA) → Horizontal Distribution Area (HDA) → Zone Distribution Area (ZDA, optional) → Equipment Distribution Area (EDA). The MDA houses the main cross-connect (MC) and typically includes core network switches, main patch panels, and LAN/SAN interconnects. In smaller data centers, MDA and HDA functions may be combined. Core routers and switches are often co-located in or near the MDA.
Question 6: What is the advantage of a 10-raised-floor tile (approximately 30-inch) raised floor height versus a 12-inch raised floor in a data center?
- A taller plenum reduces under-floor air velocity, allowing better airflow distribution and accommodating more power and data cabling (Correct answer)
- A taller raised floor provides more structural load-bearing capacity for heavy equipment
- A taller raised floor eliminates the need for perforated tiles entirely
- A taller raised floor is required by code for Tier III and above data centers
Correct answer: A taller plenum reduces under-floor air velocity, allowing better airflow distribution and accommodating more power and data cabling
A taller raised-floor plenum (typically 18–30 inches) reduces airflow velocity under the floor, improving distribution uniformity and also providing more space for power and data cabling.
Standard raised floor heights range from 12 inches to 48 inches. A taller plenum allows conditioned air to slow down and distribute more evenly before rising through perforated tiles, reducing turbulence and pressure drops across the floor. This improves the uniformity of cold air delivery to racks. Additionally, taller plenums provide more space for power whips, data cabling, and grounding conductors, which can otherwise block under-floor airflow in dense deployments. However, taller raised floors add to construction cost and building height requirements.
CDCP Data Center Design Principles 2
The Uptime Institute's Tier IV data center classification requires what minimum percentage of concurrent maintainability?