Azure · AZ-900

Cost Optimization — Azure Fundamentals (AZ-900)

18%of exam questions (21 of 114)

Committed Use, Spot, or Savings Plan: Read the Workload

The exam embeds workload descriptors — 'steady-state,' 'interruptible,' 'variable,' 'predictable' — as direct signals for which pricing model applies. Candidates over-focus on discount percentages. Azure Reservations target predictable, committed VM utilization. Savings Plans offer compute flexibility across VM families. Spot VMs serve fault-tolerant, non-time-critical jobs. Azure Cost Management provides visibility and anomaly detection — not discounts. The workload adjective in the scenario is the constraint that eliminates three of four options.

What This Pattern Tests

Azure cost optimization tests whether you match pricing models to workload patterns. Spot VMs save up to 90% for interruptible workloads. Reserved Instances (1-year or 3-year) save 30-60% for predictable compute. Azure Hybrid Benefit lets you reuse Windows Server and SQL Server licenses for up to 40% savings on VMs. The trap is recommending reservations for a workload being rightsized (you will resize it and waste the reservation) or ignoring Hybrid Benefit when the scenario mentions existing Microsoft licenses.

Decision Axis

Pricing model fit (interruptibility, commitment horizon, existing licenses) vs. raw discount percentage.

Associated Traps

Decision Rules

Which tool benchmarks on-premises CapEx/OpEx costs against Azure spend to justify migration, rather than estimating Azure-only service costs for a net-new deployment?

Azure TCO CalculatorAzure Pricing Calculator

Whether the workload's variable, event-driven utilization pattern is best served by always-on VM capacity—billed continuously regardless of use—or by a consumption-based serverless model that scales to zero and charges only per execution.

Azure FunctionsAzure Virtual Machines

Whether true scale-to-zero serverless compute (Azure Functions consumption plan) or pay-as-you-go virtual machines best satisfies the dual requirement of zero upfront CapEx and zero idle-period charges for a sporadically triggered workload.

Azure Virtual MachinesAzure Functions

Whether consumption-based serverless compute (Azure Functions) or provisioned VM-based compute (Azure Virtual Machines) satisfies the dual constraint of zero upfront CapEx and cost proportional to actual workload execution for a variable, event-driven usage pattern.

Azure FunctionsAzure Virtual Machines

Whether a workload with predictable idle periods is better served by always-on provisioned compute billed per hour regardless of load, or by consumption-based compute that scales to zero and charges only for actual execution, where eliminating idle-hour spend is the dominant constraint.

Azure FunctionsAzure Virtual Machines

When the requirement is to monitor and act on real-time or historical Azure consumption data — not estimate future costs — Azure Cost Management is the correct service; Azure Pricing Calculator operates only on hypothetical pre-deployment configurations and cannot surface or alert on live spend.

Azure Cost ManagementAzure Pricing Calculator

Choose the tool that compares existing on-premises CapEx against projected Azure OpEx (TCO Calculator) versus a tool that estimates only Azure-side resource costs (Pricing Calculator), given that the stated requirement is a before-and-after savings justification.

Azure TCO CalculatorAzure Pricing Calculator

Domain Coverage

Describe Cloud Concepts

Difficulty Breakdown

Easy: 9Hard: 12