AZ-900 Describe Azure architecture and services Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Azure Virtual Machine Scale Sets

  Why it's wrong here

  Azure Virtual Machine Scale Sets provide a layer of IaaS by deploying and automatically scaling a set of identical virtual machines, but they stop at infrastructure scaling—they include no mechanism to schedule batch tasks, track job progress, or handle retries. You'd need to add and manage your own batch orchestration logic (e.g., a custom scheduler, message queue, or an orchestrator like Kubernetes) on top of the VMs. Azure Batch, in contrast, consumes scale sets under the hood while offering the job-level management required, so a bare scale set is not sufficient.

• ✓

  Azure Batch

  Why this is correct

  Azure Batch is a managed, platform-as-a-service (PaaS) compute service built specifically for large-scale parallel and high-performance computing (HPC) batch workloads. It automatically provisions and maintains a pool of virtual machines, installs the required applications, queues and schedules tasks, monitors node health, and applies autoscaling policies based on pending tasks or fixed rules. By handling job orchestration, retry logic, and task dependencies, it relieves you from writing custom scheduling code—exactly what makes it correct for this scenario.

• ✗

  Azure Functions

  Why it's wrong here

  Azure Functions is a serverless, event-driven compute service that executes short-lived code blocks in a temporary execution environment, triggered by events such as HTTP requests, queue messages, or timers. Its consumption model imposes timeout limits (typically up to 10 minutes) and is not designed to manage persistent virtual machine pools, long-running parallel processes, or orchestrated batch pipelines. While you could use Durable Functions for some long-running workflows, Functions does not replace Azure Batch's native VM-pool management and batch scheduling, so it is wrong for this scenario.

• ✗

  Azure Container Instances

  Why it's wrong here

  Azure Container Instances (ACI) starts a single container or a small container group on demand, offering quick startup and per-second billing, but it provides no built-in batch orchestration, task queueing, or automatic VM pool scaling. You would have to implement your own scheduler and manage parallel executions or rely on external orchestration, whereas Azure Batch integrates scheduling, scaling, and failover directly into the service. Because the workload described is large-scale batch processing, ACI's lack of native batch control makes it incorrect.

Go deeper

Related to this question