AZ-104 Deploy and Manage Azure Compute Resources Practice Questions

Free AZ-104 Deploy and Manage Azure Compute Resources practice test questions

8 questions from this domain with answers and explanations - different from the samples on the main AZ-104 page. Sign up free to practice the full set.

1. Question 1Deploy and Manage Azure Compute Resources

   You have an ARM template that deploys a virtual machine to ResourceGroupA. You deploy the same template to ResourceGroupB. What happens?

   • AThe existing VM in ResourceGroupA is updated
   • BThe existing VM is moved to ResourceGroupB
   • CThe deployment fails because the VM already exists
   • DA new VM is created in ResourceGroupBCorrect
   ✓ Correct answer: D

   ARM template deployments are scoped to the resource group they target, and resource uniqueness is evaluated within that scope. Deploying the same template to ResourceGroupB creates a brand-new VM there because it has no relationship to the existing VM in ResourceGroupA. The two deployments are independent, so the original VM is unaffected. This is why the same template can be reused to stamp out resources across multiple resource groups.

   Why the other options are wrong

   • AThe existing VM in ResourceGroupA is not updated because the second deployment targets a different resource group entirely.
   • BResources are not moved by a template deployment; deploying to ResourceGroupB does not relocate the VM from ResourceGroupA.
   • CThe deployment does not fail; the VM name only needs to be unique within ResourceGroupB, where no such VM exists yet.
2. Question 2Deploy and Manage Azure Compute ResourcesSelect all that apply

   You are planning to deploy Azure Virtual Machines for a production workload that requires high availability. Which of the following configurations help achieve higher availability for Azure VMs? (Choose two.)

   • APlace VMs in the same Availability Set with multiple fault domains and update domainsCorrect
   • BDeploy all VMs on the same physical host using a Dedicated Host
   • CUse the same storage account for all VM disks
   • DDeploy VMs across multiple Availability Zones within the same regionCorrect
   ✓ Correct answer: A, D

   Availability Sets distribute VMs across multiple fault domains (separate physical racks) and update domains (logical groups for planned maintenance), providing protection against hardware failures and planned maintenance events within a single datacenter with a 99.95% SLA. Availability Zones distribute VMs across physically separate datacenters within a region, each with independent power, cooling, and networking, providing a higher 99.99% SLA and protection against entire datacenter failures. Both configurations are core Azure high availability strategies.

   Why the other options are wrong

   • BDeploying all VMs on the same physical host using a Dedicated Host creates a single point of failure because a hardware issue on that host would affect all VMs simultaneously.
   • CUsing the same storage account for all VM disks does not improve availability and could create a performance bottleneck or single point of failure for storage-related issues.
3. Question 3Deploy and Manage Azure Compute ResourcesSelect all that apply

   Litware Inc needs to ensure high availability for a critical application running on Azure virtual machines. The application must be protected against both hardware failures within a datacenter and datacenter-level failures within a region. Which two configurations should you implement? (Choose two.)

   • ADeploy VMs in an availability set with multiple fault domainsCorrect
   • BEnable boot diagnostics on all VMs
   • CDeploy VMs across multiple availability zonesCorrect
   • DConfigure Azure Site Recovery for the VMs
   ✓ Correct answer: A, C

   Availability sets protect against hardware failures within a datacenter by distributing VMs across multiple fault domains and update domains, ensuring that a hardware rack failure or planned maintenance affects only a subset of your VMs. Availability zones protect against datacenter-level failures by placing VMs in physically separate datacenter facilities within the same Azure region, each with independent power, cooling, and networking. Together, these configurations address both hardware-level and datacenter-level failures.

   Why the other options are wrong

   • BEnabling boot diagnostics provides screenshots and serial console logs for troubleshooting VM startup issues but does not contribute to high availability.
   • DConfiguring Azure Site Recovery is a disaster recovery solution for replicating VMs to another region; it does not provide within-region high availability.
4. Question 4Deploy and Manage Azure Compute Resources

   Munson's Pickles and Preserves Farm is deploying a multi-container application using Docker Compose. The application consists of a web frontend, an API backend, and a Redis cache. You need a fully managed solution that supports Docker Compose files directly. What should you use?

   • AAzure Kubernetes Service with a custom Helm chart
   • BAzure App Service for Containers with single-container deployment
   • CAzure Virtual Machines with Docker installed
   • DAzure Container Instances with a multi-container group using a YAML fileCorrect
   ✓ Correct answer: D

   Azure Container Instances is a fully managed, serverless container service that supports multi-container groups defined in a YAML file, conceptually similar to Docker Compose. A single container group can run the web frontend, API backend, and Redis cache together on shared resources with no underlying infrastructure to manage, making it the fitting fully managed solution for this scenario.

   Why the other options are wrong

   • AAzure Kubernetes Service with a custom Helm chart delivers full orchestration but requires managing cluster nodes and Helm packaging, which is far more complex than this small multi-container app needs and is not the lightweight managed option requested.
   • BAzure App Service for Containers with single-container deployment runs only one container per app on the standard tier and does not natively run a Docker Compose style multi-container group of separate frontend, backend, and cache containers.
   • CAzure Virtual Machines with Docker installed forces you to manage the OS, patching, and scaling of the host, which violates the requirement for a fully managed solution.
5. Question 5Deploy and Manage Azure Compute Resources

   You have deployed a web application to Azure App Service and need to configure automatic scaling based on HTTP queue length. Which scaling type should you configure?

   • AScale out (horizontal scaling) with a custom autoscale rule based on HTTP queue length metricCorrect
   • BManual scaling to a fixed number of instances sized for expected peak load
   • CScale up (vertical scaling) to a larger instance size when the queue grows
   • DScale up by switching the plan to a higher Premium v3 pricing tier
   ✓ Correct answer: A

   Azure App Service supports custom autoscale rules that monitor specific metrics such as HTTP queue length. When the queue grows beyond a defined threshold, the autoscale engine automatically adds instances to handle the increased load. This approach is fully automatic and responds dynamically to changing demand patterns.

   Why the other options are wrong

   • BManual scaling sets a static instance count that never reacts to the HTTP queue length metric, so it cannot automatically respond to changing request backlog.
   • CScaling up increases the size of each instance but does not add instances; App Service autoscale that responds to a metric like HTTP queue length works by scaling out horizontally, not by resizing.
   • DMoving to a higher pricing tier is a manual vertical change that raises capacity per instance; it is not an automatic response to the HTTP queue length metric and does not add instances dynamically.
6. Question 6Deploy and Manage Azure Compute Resources

   You have the inbound NSG rules shown in the table associated with Web-Subnet. An admin from IP 10.0.1.5 tries to SSH (port 22) to a VM in Web-Subnet. What happens?

   • AThe connection is allowed by AllowSSH
   • BThe connection times out
   • CThe connection is denied by DenyAllCorrect
   • DThe connection is allowed by a default NSG rule
   ✓ Correct answer: C

   NSG rules are processed in priority order from the lowest number to the highest, and the first matching rule decides the outcome. Although the AllowSSH rule (priority 300) would match this traffic by source (10.0.1.5 is within 10.0.1.0/24) and port (22), it is never reached because the catch-all DenyAll rule (priority 200) is evaluated first and matches the traffic. The connection is therefore denied by the DenyAll rule.

   Why the other options are wrong

   • AAllowSSH does match this traffic by source and port, but it is at priority 300 and is never evaluated because DenyAll at priority 200 matches first; the lower-numbered DenyAll wins.
   • BAn NSG denial drops the packet per the matching rule; the deciding factor here is the explicit DenyAll match at priority 200, not an absence of any matching rule.
   • DA default rule does not decide this case; the explicit DenyAll rule at priority 200 matches before any default rule would apply.
7. Question 7Deploy and Manage Azure Compute ResourcesSelect all that apply

   Contoso Manufacturing operates a fleet of Windows Server 2022 virtual machines in Azure that run a real-time inventory tracking application. The operations team has reported that during peak warehouse hours, the application experiences intermittent high-latency disk I/O, causing order processing delays. The current VMs use Standard SSD managed disks. The team needs to improve disk performance while also ensuring that the VMs can sustain sub-millisecond latency for their most critical database workloads. Requirements: - The database tier VMs require consistent sub-millisecond storage latency - The application tier VMs need better throughput than Standard SSD but do not require sub-millisecond latency - The solution must use managed disks - Cost optimization is a secondary concern to performance Which TWO disk types should you recommend? (Choose two.)

   • APremium SSD managed disks for the application tier VMsCorrect
   • BStandard HDD managed disks for the application tier VMs
   • CUltra Disk managed disks for the database tier VMsCorrect
   • DPremium SSD v2 managed disks with adjusted IOPS for the database tier VMs
   ✓ Correct answer: A, C

   Premium SSD managed disks deliver significantly higher IOPS and throughput compared to Standard SSD managed disks, providing the improved disk performance that the application tier VMs need for inventory tracking operations without requiring sub-millisecond latency. Ultra Disk managed disks are purpose-built for the most demanding data-intensive workloads and deliver consistent sub-millisecond latency, making them the correct choice for the database tier VMs that require guaranteed sub-millisecond storage latency for real-time processing. Both disk types are managed disks, meeting the stated requirement.

   Why the other options are wrong

   • BStandard HDD managed disks for the application tier VMs is incorrect because Standard HDD provides lower performance than the current Standard SSD disks, which would degrade rather than improve the application's disk I/O performance.
   • DPremium SSD v2 managed disks with adjusted IOPS for the database tier VMs is incorrect because while Premium SSD v2 also offers sub-millisecond latency and granular performance tuning, the question asks for two answers and Ultra Disk is the established choice specifically designed for database workloads requiring guaranteed sub-millisecond latency with the highest possible throughput.
8. Question 8Deploy and Manage Azure Compute Resources

   You have the managed disks shown in the table. You need to attach DataDisk-2 to VM-Web1. The disk and VM are in the same region and availability zone. What should you verify before attaching the disk?

   • AThat DataDisk-2 uses the same encryption type as OSDisk-VM1
   • BThat VM-Web1 supports the Standard SSD disk type
   • CThat DataDisk-2 is converted to Premium SSD first
   • DThat the VM size supports the total number of data disks after attachingCorrect
   ✓ Correct answer: D

   Every Azure VM size enforces a maximum number of attachable data disks, and that ceiling is the real constraint when adding a disk. VM-Web1 already has DataDisk-1 attached, so attaching DataDisk-2 means the VM size must permit at least two data disks or the attach call will be rejected. Verifying the data disk count against the VM size limit is therefore the meaningful precheck.

   Why the other options are wrong

   • AMatching the encryption type to OSDisk-VM1 is unnecessary because data disks and OS disks can use different encryption (platform-managed versus customer-managed keys) on the same VM.
   • BConfirming Standard SSD support is pointless because every Azure VM size supports Standard SSD data disks regardless of the VM family or size.
   • CConverting DataDisk-2 to Premium SSD first is not required, since a single VM can attach a mix of Standard and Premium disks with no uniformity requirement.

How Deploy and Manage Azure Compute Resources is tested

This domain holds 200 of the 956 questions in the AZ-104 bank, about 21%. The mix is 112 single-answer multiple choice, 23 multiple-response, 20 true/false, 20 yes/no scenario, 18 hotspot and 7 ordering, so it is worth practising the formats as well as the content.

Once you have a few attempts recorded, CertGrid scores every domain separately and points you at the weakest one, so you can drill Deploy and Manage Azure Compute Resources on its own rather than re-running full-length mocks.

Other AZ-104 exam domains

• Manage Azure Identities and Governance225 questions
• Implement and Manage Storage195 questions
• Implement and Manage Virtual Networking194 questions
• Monitor and Maintain Azure Resources142 questions
• All AZ-104 practice questions956 total
• Deploy and Manage Azure Compute Resources study notesKey concepts
• Microsoft practice examsAll Microsoft

AZ-104 Deploy and Manage Azure Compute Resources FAQ