Operator Academy
A field guide to the systems behind a data center. Walk the power room, cooling plant, rack hall and operations center; learn the language; then test each idea against the facility you run in Data Center Fan.
Explore 8 visual field lessons, interactive concepts and 51 defined terms. Each lesson links real facility practice to the systems simulated in Data Center Fan.
PUE and Facility Efficiency
Follow every kilowatt from the utility meter to useful compute and learn what the ratio can—and cannot—tell an operator.
- Calculate PUE from facility and IT power
- Identify the correct measurement boundary
- Separate infrastructure efficiency from IT productivity
From Utility to Rack: The Power Chain
Map the electrical path that keeps a server online and find the devices that condition, switch, distribute or back up the load.
- Name the major stages from utility to rack
- Distinguish stored energy from generation
- Recognize a single point of failure
Redundancy: N, N+1 and 2N
Build enough spare capacity to survive the failure you actually designed for—without confusing extra boxes with independent paths.
- Express required capacity as N
- Compare N+1 with fully duplicated 2N paths
- Test a design through maintenance and failure scenarios
Hot Aisle, Cold Aisle and Airflow
Cooling is a path: deliver conditioned air to equipment inlets and return hot exhaust without letting the two streams short-circuit.
- Orient racks into hot and cold aisles
- Recognize bypass and recirculation
- Explain why containment can reduce fan and cooling work
CRAC, CRAH, Chillers and Liquid Cooling
Trace heat from silicon to the outside environment and match cooling architecture to density, climate, water and operational constraints.
- Differentiate CRAC and CRAH heat paths
- Follow chilled and condenser water loops
- Recognize when liquid cooling shortens the thermal path
SLA, Availability, MTTR and Fault Domains
Reliability is a service outcome: define what must remain available, how failure is isolated and how quickly operators can restore it.
- Convert an SLA percentage to a downtime budget
- Explain the roles of MTBF and MTTR
- Draw a useful fault-domain boundary
Rack Units, Power Density and Capacity
A rack is full when its first binding constraint is exhausted—not necessarily when the last rack unit is occupied.
- Read rack-unit notation
- Calculate average rack power density
- Identify the first binding capacity constraint
DCIM, BMS, Alarms and Change Control
Instrumentation is useful only when operators trust the signal, know the owner and can act through a controlled path.
- Separate BMS, EPMS, DCIM and IT monitoring roles
- Design an actionable alarm
- Use change control without turning it into paperwork theatre