AZ-305 domain - 38% of the exam

Design Infrastructure Solutions

Design Infrastructure Solutions is 38% of the AZ-305 exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

98
Practice questions
4
Objectives
65
Medium
33
Hard

Where people go wrong

  • Giving the legacy Consumption plan an unbounded timeout

    The wrong answers attach an unbounded execution timeout to the legacy Consumption plan, or give that plan an Always On setting. On Consumption the documented maximum is 10 minutes, with a 5-minute default, and Always On is not exposed. Unbounded execution is documented for Flex Consumption, Premium, Dedicated and Container Apps. On the Dedicated plan it applies only once Always On is enabled on the plan that hosts the function app.

    See it in a sample question
  • Expecting Container Apps to expose the Kubernetes API

    The wrong answers pick Container Apps, Container Instances or Functions on Container Apps when application teams need direct Kubernetes API and control-plane access. Container Apps is built on Kubernetes but hides the API, and Container Instances runs single container groups with no orchestrator. Azure Kubernetes Service is where the decision tree routes teams that need that access, Helm charts, or a cluster of up to 5,000 nodes on the Standard or Premium tier.

    See it in a sample question
  • Treating Traffic Manager as a proxy

    The wrong answers expect Traffic Manager to rewrite headers, route by URL path or fail over in-flight requests. Traffic Manager works only at the DNS level: once it returns a DNS response the client connects directly to the endpoint, and DNS TTL caching slows failover. Header transformation, TLS termination at the edge and re-routing on origin failure belong to Front Door, a global Layer 7 proxy, while path-based routing inside one region belongs to Application Gateway.

    See it in a sample question
  • Confusing Service Endpoints with Private Endpoints

    The wrong answers credit a Service Endpoint with what only a Private Endpoint does. A Service Endpoint works from inside the VNet and keeps the service's public IP and FQDN, so on-premises clients cannot use it. A Private Endpoint maps one resource instance to a private IP from the VNet, protects against data exfiltration, and is reachable over ExpressRoute or VPN. Clients resolve that private IP only when the Azure Private DNS zone for the service, such as the one for blob.core.windows.net, is linked to their VNet.

    See it in a sample question
  • Misplacing features between Classic Premium and Premium v2

    The wrong answers misplace features between the two Premium tiers of API Management. An upgrade to Premium v2 is assumed to bring multi-region deployment, yet multi-region deployment under a single instance is supported only on Classic Premium, and so is the self-hosted gateway. In the other direction, availability zones, workspaces and the 5 GB built-in cache are treated as Classic Premium exclusives, when Premium v2 has them too. Standard v2 and Consumption have no multi-region deployment either.

    See it in a sample question

Objectives in this domain

Sample questions from this domain

Free sampleDesign Infrastructure Solutionsmedium

A team migrates a 12-minute end-of-day report function from on-premises to Azure Functions. The current draft uses the legacy Consumption plan. What is the documented maximum function timeout on that plan?

  • AUnbounded execution duration up to 60 minutes on the Consumption plan.
  • B10 minutes maximum on the Consumption plan, with a 5-minute default. Correct
  • C30 minutes maximum on the Consumption plan, matching other serverless plans.
  • D230 seconds maximum on any HTTP-triggered Function regardless of plan.
Legacy Consumption tops out at 10 minutes per execution; everything longer needs Flex Consumption, Premium, or Dedicated. The Functions hosting comparison table lists the legacy Consumption plan with a 5-minute default and 10-minute maximum timeout. A 12-minute job will not fit and the team should choose Flex Consumption, Premium, or Dedicated for an unbounded timeout.

Why A is wrong: Unbounded execution is documented for Flex Consumption, Premium, Dedicated, and Container Apps, not the legacy Consumption plan.

Why B is correct: Correct. The Functions hosting comparison table lists the legacy Consumption plan with a 5-minute default and 10-minute maximum timeout.

Why C is wrong: 30 minutes is the default timeout on the other plans, not the maximum on Consumption.

Why D is wrong: 230 seconds is the HTTP-trigger response limit imposed by the Azure Load Balancer; it is not the overall function timeout for non-HTTP triggers.

Free sampleDesign Infrastructure Solutionsmedium

A platform engineering group runs a single regional cluster that must scale to 5,000 worker nodes for a high-throughput batch and inference workload. The cluster must remain on the Standard or Premium tier, support pod and cluster autoscaling, and continue to expose the Kubernetes API to its application teams. Which Azure container service does the decision tree route this workload to?

  • AAzure Container Apps with replica scaling rules
  • BAzure Kubernetes Service on the Standard or Premium tier Correct
  • CAzure Container Instances container groups
  • DAzure Red Hat OpenShift managed cluster
A 5,000-node ceiling with direct Kubernetes API access is the documented AKS signature on the Standard or Premium tier. The AKS overview and the compute decision tree both record the 5,000-node-per-cluster scale limit for AKS on the Standard or Premium tier. The other container services have substantially lower documented scale ceilings or solve a different shape of problem entirely. Container Apps tops out at 1,000 replicas per revision and does not expose the Kubernetes API. ACI is bounded at 100 container groups per subscription by default. Azure Red Hat OpenShift is a managed OpenShift cluster aimed at OpenShift-invested teams, not the documented routing target for a generic 5,000-node Kubernetes API requirement.

Why A is wrong: Container Apps is bounded at 1,000 replicas per revision per environment and does not expose the Kubernetes API, so it fails both the scale and API requirements.

Why B is correct: Correct. The AKS overview and the compute decision tree both record the 5,000-node-per-cluster scale limit for AKS on the Standard or Premium tier.

Why C is wrong: Container Instances runs single container groups without orchestration and has a default ceiling of 100 container groups per subscription, far below 5,000 nodes.

Why D is wrong: Azure Red Hat OpenShift is positioned for teams with an existing OpenShift investment; the compute decision tree does not route a generic 5,000-node Kubernetes API workload to it.

Free sampleDesign Infrastructure Solutionsmedium

An architect explains that after a Traffic Manager DNS response, all subsequent client traffic is proxied through Traffic Manager so the service can apply WAF rules and request rewrites. Traffic Manager sits in the data path between client and endpoint after the initial DNS resolution. Is this statement correct?

  • AYes
  • BNo Correct
Traffic Manager is DNS-only and not in the data path; HTTP inspection is Front Door or Application Gateway territory. Documentation states Traffic Manager works only at the DNS level: once the DNS response is returned, Traffic Manager is no longer involved and the client connects directly to the endpoint. It cannot apply WAF or rewrite requests because it never inspects traffic.

Why A is wrong: Treating Traffic Manager as a proxy leads designers to expect features that belong to Front Door or Application Gateway, which actually terminate and inspect HTTP.

Why B is correct: Correct. Documentation states Traffic Manager works only at the DNS level: once the DNS response is returned, Traffic Manager is no longer involved and the client connects directly to the endpoint.

Free sampleDesign Infrastructure Solutionsmedium

An auditor requires that an Azure SQL Database be reachable from an on-premises office over an ExpressRoute private peering circuit and that no public internet route to the database be possible. Which capability satisfies both requirements?

  • AEnable a Microsoft.Sql Service Endpoint on the on-premises gateway subnet of the hub VNet.
  • BAdd the on-premises subnet to the SQL Database firewall allow-list using a public IP rule.
  • CEnable the SQL Database 'Allow Azure services' setting with VNet integration on the App Service.
  • DDeploy a Private Endpoint for the SQL Database in the hub VNet with private DNS zone integration. Correct
Private Endpoint is the only choice that supports on-premises access via ExpressRoute or VPN while removing public-internet reachability. Private Endpoint maps the SQL Database to a private IP from the VNet and is reachable from on-premises through ExpressRoute private peering or VPN. Disabling public network access on the database and configuring the Azure private DNS zone removes the public path entirely.

Why A is wrong: Service Endpoints only work from inside the VNet; they cannot be consumed from on-premises locations and they still use the public IP of the service.

Why B is wrong: Firewall allow-listing leaves the SQL Database publicly reachable on the internet for any source not blocked, which violates the requirement that no public route exist.

Why C is wrong: The Allow Azure services toggle exposes the database to all Azure tenants and does not address on-premises access or remove the public endpoint.

Why D is correct: Correct. Private Endpoint maps the SQL Database to a private IP from the VNet and is reachable from on-premises through ExpressRoute private peering or VPN.

Free sampleDesign Infrastructure Solutionshard

A global API platform must run gateways in three Azure regions under a single API Management instance with active-active multi-region routing. Which tier supports this?

  • AUse the Premium v2 tier and add regional gateways during provisioning.
  • BUse the Standard v2 tier with outbound VNet integration in three regions.
  • CUse the Classic Premium tier for the multi-region API Management deployment. Correct
  • DUse the Consumption tier and rely on automatic regional scaling.
Multi-region API Management deployment is a Classic Premium exclusive; Premium v2 does not support it. The feature table explicitly lists multi-region deployment as supported only on Classic Premium. Premium v2 adds availability zones and 30 scale units but does not include multi-region; this is one of the exam's key tier differentiators.

Why A is wrong: Premium v2 supports availability zones and 30 scale units but does not offer multi-region deployment; that capability is reserved for Classic Premium.

Why B is wrong: Standard v2 offers outbound VNet integration and workspaces but cannot host a single multi-region API Management service.

Why C is correct: Correct. The feature table explicitly lists multi-region deployment as supported only on Classic Premium.

Why D is wrong: Consumption is a serverless gateway with no multi-region deployment, no static IP, and no built-in cache; it cannot satisfy the requirement.

Other domains in this exam

See also the AZ-305 cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by Microsoft. Original, blueprint-aligned practice material only.