This chapter covers Azure Reserved Instances, a pricing option for workloads with predictable, long-term usage patterns. AZ-900 tests this under objective 2.2, alongside other compute pricing options.
Jump to a section
A simple way to picture Azure Reserved Instances
A store might offer a discount if you commit to buying a certain item regularly — say, 20% off if you pre-commit to buying 100 items a month for the next year. Azure Reserved Instances work similarly: commit to using a virtual machine for a fixed period (1 or 3 years), and get a significant discount off the standard pricing in exchange.
What Reserved Instances are
Reserved Instances are a pricing option where you pay upfront (or over a fixed period) for a 1-year or 3-year commitment to use a specific VM size in a specific region, and receive a significant discount (typically 30-70% off standard pricing) in exchange.
Who they fit
Reserved Instances fit organizations with predictable, long-term workloads — for example, a steady-state production application that's always running. The longer the commitment and the more upfront you pay, the bigger the discount.
The tradeoff
You save money if your usage patterns match the commitment, but you're locked into that commitment — if your needs change or you no longer need the resources, you've already paid and might lose money compared to pay-as-you-go.
How they compare to other options
Spot Instances offer the biggest discounts but with the risk of interruption. Reserved Instances offer a smaller discount without interruption risk, requiring a long-term commitment instead. Standard, pay-as-you-go instances have no discount but maximum flexibility.
Confirm long-term, predictable usage
Reserved Instances only make sense for workloads you're confident will run consistently for the full commitment period.
Choose commitment length
1-year or 3-year commitments are available — longer commitments typically offer bigger discounts.
Choose payment option
Upfront, partial, or spread-over-time payment options affect the discount — upfront payment generally means the biggest discount.
Reserve the appropriate capacity
Reserve the specific VM size and region matching your expected needs.
A company with a steady-state production environment that runs 24/7 purchases 3-year Reserved Instances for those VMs, saving 60% compared to pay-as-you-go pricing, confident the production environment will continue for years.
Objective 2.2 expects candidates to understand Reserved Instances as a cost-saving option for long-term, predictable workloads, alongside other pricing options.
A common wrong answer is assuming Reserved Instances are always the right choice — they only save money if your actual usage matches the commitment. Unexpected downsizing or workload migration can make them a poor choice.
Stable terms: Reserved Instances. Memory trick: Reserved Instances = big discount (30-70%) for long-term commitment — great if usage is predictable, risky if needs might change.
Azure Reserved Instances offer significant discounts (typically 30-70%) for committing to 1 or 3-year usage of specific VM sizes.
They're suitable for long-term, predictable workloads with stable capacity needs.
The longer the commitment and more upfront the payment, the bigger the discount.
They're less suitable for workloads with uncertain or fluctuating long-term demands.
These come up on the exam all the time. Here's how to tell them apart.
Pay-as-You-Go
No upfront commitment
Maximum flexibility
Standard pricing, no discount
Best for unpredictable or short-term usage
Reserved Instances
1 or 3-year commitment required
Significant discount (30-70%)
Best for predictable, long-term workloads
Mistake
Reserved Instances are always cheaper than pay-as-you-go.
Correct
They're cheaper only if your actual usage matches the commitment — if your needs change, you could end up paying more.
Mistake
Reserved Instances are suitable for all workloads.
Correct
They're most suitable for long-term, predictable workloads with stable capacity needs — workloads with fluctuating or uncertain long-term demand are better suited to pay-as-you-go or Spot Instances.
They're a pricing option where you commit to using a specific VM size in a specific region for 1 or 3 years, receiving a significant discount (typically 30-70% off) in exchange for that commitment.
Only if your actual usage matches the commitment — if your needs change and you no longer need the reserved capacity, you could end up paying more than if you'd used pay-as-you-go instead.
Long-term, predictable workloads with stable capacity needs — for example, a steady-state production application expected to run for years with consistent resource requirements.
You've just covered Azure Reserved Instances — now see how well it sticks with free AZ-900 practice questions. Full explanations included, no account needed.
Done with this chapter?