A logistics company is designing the compute infrastructure for a new real-time fleet tracking application. The application consists of a stateless API layer that runs continuously 24/7. The API layer requires a 99.99% availability SLA and must be deployed across multiple Availability Zones to ensure resiliency. The workload is highly predictable with a steady-state CPU and memory usage profile. The company wants to minimize compute costs over a three-year period without risking service interruption.
Which compute solution should you recommend to meet these requirements?
- ADeploy the API layer on Azure Spot Virtual Machines across three Availability Zones.
- Deploy the API layer on standard Azure Virtual Machines configured with three-year Azure Reserved Virtual Machine Instances across multiple Availability Zones.Cevap
- CDeploy the API layer on Azure Virtual Machine Scale Sets in a single Availability Zone using Pay-As-You-Go D-series virtual machines.
- DDeploy the API layer on Azure Virtual Machine Scale Sets using the Uniform orchestration mode with Spot priority across multiple Availability Zones.
Cevap
Deploy the API layer on standard Azure Virtual Machines configured with three-year Azure Reserved Virtual Machine Instances across multiple Availability Zones.
Deploying standard virtual machines configured with three-year Azure Reserved Virtual Machine Instances across multiple Availability Zones is correct. Reserved instances provide significant cost savings (up to 72% compared to pay-as-you-go) for predictable, steady-state workloads with a long-term commitment. Deploying across multiple zones meets the 99.99% availability SLA without the risk of eviction.
Adım Adım Çözüm
Anahtar Kavram
Selecting Azure VM purchasing models and resiliency configurations based on SLA, cost constraints, and workload predictability.