I85PREP
⌘ K
CDFOM study materials
CDFOM study resources

CDFOM: Capacity, Availability and Continuity Management: Redundancy, RTO and RPO

CDFOM · Work through your exam, one topic at a time.

All guide sections

Capacity, Availability and Continuity Management: Redundancy, RTO and RPO

Capacity, Availability and Continuity Management

This domain tests forward planning and resilience: power, cooling, space and network capacity; availability design; and emergency and continuity response. The exam expects calculations that respect redundancy, not just nameplate capacity.

Core rules and concepts

  • Capacity planning covers power (UPS, PDU, generator), cooling (CRAH, CRAC, chilled water), space (racks, floor tiles), network (ports, fibre, bandwidth) and weight. Track peak, not average, and forecast from growth and pipeline.
  • Redundancy models include N, N+1, 2N, concurrent maintainability and fault tolerance. A redundant system can only carry its design load with the redundant path available; derating matters.
  • Availability is measured against agreed boundaries; design availability depends on single points of failure, maintenance capability and MTTR.
  • Continuity includes BCP, DR, RTO for time to restore, RPO for data loss tolerance, crisis management, escalation and emergency response. Plans need testing, lessons learned and update.
  • Emergency response covers fire, power loss, cooling loss, security breach, flood and utility failure, with defined roles and communication.

Worked example

A data hall has two 500 kVA UPS modules configured N+1, supporting a critical load of 420 kVA. Total installed capacity is 1,000 kVA, but with one module out the remaining 500 kVA must carry the load. Utilization in redundant mode = 420 ÷ 500 = 84%, so the site is within capacity but has limited headroom. If the load grows to 520 kVA, the N+1 configuration can no longer survive a single UPS failure; the site is effectively N for that load. The correct conclusion is to add capacity or reduce load before accepting new IT. Reasoning: test capacity against the worst-case redundant state, not the total installed rating.

Traps that cost marks

  • Reporting average utilization and missing peak demand.
  • Adding load without checking the redundancy state.
  • Confusing RTO (restore time) with RPO (data loss).
  • Treating an untested DR plan as continuity capability.
  • Ignoring cooling, floor loading and network capacity when adding power capacity.