Data Center Fundamentals and Infrastructure
A data center is best understood as a system of interdependent subsystems: space and structure (raised floor or slab, racks, containment), power path, cooling path, network path, and management/security layers. Candidates should know the difference between the white space (the IT floor) and grey space (support areas such as electrical rooms and mechanical plants), and between the three classic availability tiers (roughly: single-path no redundancy, partial redundancy, and concurrently maintainable with multiple distribution paths).
Key concepts to master: redundancy notation (N, N+1, 2N, 2(N+1)) and what each implies about fault tolerance versus concurrent maintainability; the difference between availability and fault tolerance; hot/cold aisle layout and airflow containment; and the distinction between design topology and actual operational redundancy.
Common traps: confusing N+1 with 2N (N+1 tolerates one component failure but may not allow maintenance without risk); assuming a UPS makes the whole site redundant when the cooling plant is single-path; and treating tier level as a guarantee rather than a design intent. Another trap is mixing up kW capacity, power density per rack, and total IT load.
Worked example / rule of thumb: if a row of 10 racks each draws 6 kW, the row's IT load is 60 kW; size the power and cooling distribution for that plus headroom, not for the nameplate capacity of every power supply. Rule of thumb: redundancy protects against failure, topology protects against maintenance - you need both for high availability.