I85PREP
⌘ K
IBM Cloud Technical Advocate study materials
IBM Cloud Technical Advocate study resources

IBM Cloud Technical Advocate practice questions

One question at a time. Build your IBM Cloud Technical Advocate understanding.

First 50 of 200 questions. Each one lists the choices, the correct answer, and an explanation. The checker at the top of this page saves progress in this browser when it is available.

Continue in the practice tool above to load more questions, 50 at a time.

Question 1Cloud Computing Concepts and Business Value

A developer uses the IBM Cloud console to create an Object Storage bucket immediately, without opening a support ticket or waiting for an administrator. Which essential cloud characteristic does this primarily demonstrate?

  1. Resource pooling
  2. Broad network access
  3. Rapid elasticity
  4. On-demand self-service Correct answer.

Explanation

On-demand self-service is the cloud characteristic in which a consumer can unilaterally provision computing capabilities, such as storage, as needed automatically without requiring human interaction with the provider. Creating a bucket from a console without a ticket is exactly that behavior. Broad network access concerns reachability over standard network mechanisms, while pooling and elasticity concern shared resources and scaling.

Question 2Cloud Computing Concepts and Business Value

A CFO wants to move spending away from upfront data center hardware purchases and toward a usage-based monthly cloud bill. Which business value does this shift describe?

  1. Disaster recovery improvement
  2. CAPEX-to-OPEX shift Correct answer.
  3. Vendor lock-in increase
  4. Economies of scale

Explanation

Cloud consumption converts many large upfront capital expenditures into recurring operating expenses tied to usage. This improves cash-flow flexibility and avoids overprovisioning for peak capacity. Economies of scale can lower unit prices, but the specific change from upfront purchase to monthly usage billing is the CAPEX-to-OPEX shift.

Question 3Cloud Computing Concepts and Business Value

Under the shared responsibility model for an IBM Cloud IaaS virtual server, which task remains the customer's responsibility?

  1. Hypervisor patching
  2. Network fabric maintenance
  3. Physical security of the data center
  4. Guest operating system patching Correct answer.

Explanation

In IaaS, the provider secures the facilities, hardware, network fabric, and hypervisor, while the customer manages the guest operating system and everything installed above it. Patching the guest OS therefore belongs to the customer. Physical security, hypervisor patching, and network fabric maintenance remain with the cloud provider.

Question 4Cloud Computing Concepts and Business Value

A workload automatically adds application instances when CPU utilization exceeds 70 percent and removes them when utilization drops. Which cloud property is being demonstrated?

  1. Multi-tenancy
  2. Fault tolerance
  3. Rapid elasticity Correct answer.
  4. Vertical scaling

Explanation

Rapid elasticity is the ability to scale resources outward and inward automatically to match demand. Adding and removing instances in response to CPU thresholds is horizontal scaling driven by demand. Vertical scaling would resize a single server, multi-tenancy concerns shared provider resources, and fault tolerance concerns continued operation despite failures.

Question 5Cloud Computing Concepts and Business Value

A development team wants to deploy containerized code without managing servers, clusters, or operating system patches, and wants to pay only when requests are processed. Which IBM Cloud offering model fits this requirement?

  1. SaaS CRM with per-user licensing
  2. Serverless (IBM Cloud Code Engine) Correct answer.
  3. IaaS virtual servers with customer-managed OS
  4. Bare metal servers with dedicated hosts

Explanation

Serverless platforms such as IBM Cloud Code Engine run containerized code while the provider manages servers, clusters, and underlying patches. Billing is based on actual usage, such as requests or resource consumption while code runs. IaaS and bare metal still require the customer to manage operating systems or servers, and SaaS delivers a finished application rather than a deployment target for custom code.

Question 6Cloud Computing Concepts and Business Value

Which statement best explains why a public cloud provider can often charge less per unit of compute than many private data centers?

  1. Every tenant receives dedicated physical hosts by default
  2. Workloads avoid network virtualization entirely
  3. Resources are pooled across tenants to raise utilization Correct answer.
  4. Customers purchase and own the underlying hardware

Explanation

Public cloud providers pool resources across many tenants and use high utilization and economies of scale to lower the cost per unit. Customers consume shared capacity rather than paying for dedicated hardware that may sit idle. Dedicated hosts, customer-owned hardware, and avoiding virtualization describe private or less efficient models, not the source of public cloud cost advantage.

Question 7Cloud Computing Concepts and Business Value

A business requires that after a regional failure, service must resume within 15 minutes and no more than 5 minutes of transactions may be lost. Which pair correctly describes these requirements?

  1. RTO of 5 minutes and RPO of 15 minutes
  2. RTO of 15 minutes and RPO of 5 minutes Correct answer.
  3. MTBF of 15 minutes and RTO of 5 minutes
  4. SLA of 15 minutes and MTTR of 5 minutes

Explanation

Recovery time objective, or RTO, is the target time to restore service after a disruption, so 15 minutes is the RTO. Recovery point objective, or RPO, is the maximum acceptable amount of data loss measured in time, so 5 minutes is the RPO. SLA and MTTR describe service commitments and repair time, while MTBF describes reliability between failures.

Question 8Cloud Computing Concepts and Business Value

When comparing a simple on-premises server purchase with a public cloud total cost of ownership, which cost category is most often omitted from the on-premises side?

  1. Power, cooling, and facilities costs Correct answer.
  2. The storage array list price
  3. The software license list price
  4. The published purchase price of the servers

Explanation

A fair total cost of ownership comparison must include data center power, cooling, floor space, and facilities operations, which are easy to overlook when focusing on hardware list prices. Cloud bills bundle these costs into service rates, so omitting them makes on-premises appear cheaper than it is. Server, storage, and software license prices are usually visible in a purchase quote and are therefore less likely to be omitted.

Question 9Cloud Computing Concepts and Business Value

An enterprise must exit a data center within 90 days with minimal application changes. Which migration strategy best fits this constraint while limiting cloud-native benefits?

  1. Rehost (lift-and-shift) Correct answer.
  2. Rearchitect as serverless functions
  3. Retire and replace with SaaS
  4. Refactor into microservices

Explanation

Rehosting, also called lift-and-shift, moves workloads to cloud infrastructure with little or no code change, which supports aggressive migration timelines. It preserves the existing application architecture, so it captures infrastructure benefits but limits cloud-native gains such as microservices resilience or serverless scaling. Refactoring, rearchitecting, and replacing with SaaS require more design and change effort than a 90-day minimal-change exit typically allows.

Question 10Cloud Computing Concepts and Business Value

A team runs serverless workloads on IBM Cloud Code Engine and wants to reduce lock-in while keeping managed serverless operations. Which approach best balances portability and managed benefits?

  1. Package functions as containers with portable APIs Correct answer.
  2. Avoid containers and rely on vendor CLI scripts
  3. Move all state into a proprietary managed database
  4. Use provider-specific proprietary event triggers for every integration

Explanation

Packaging functions as container images and exposing portable APIs keeps the application deployable on other container-capable platforms while still using the provider's managed serverless runtime. Lock-in increases when integrations, state, or automation depend on proprietary services and interfaces. A balanced approach isolates provider-specific dependencies and documents them so migration remains feasible.

Question 11Cloud Computing Concepts and Business Value

A hospital keeps its electronic health records in its own data center for regulatory reasons and runs a new analytics platform in IBM Cloud VPC. A single identity provider authenticates staff to both environments, and some batch workloads move between them on a schedule. Which cloud deployment model does this describe?

  1. Private cloud
  2. Hybrid cloud Correct answer.
  3. Community cloud
  4. Public cloud

Explanation

Hybrid cloud combines a private or on-premises environment with public cloud services and connects them through identity, networking, and workload orchestration. The shared identity provider and scheduled movement of workloads across both environments are the defining signals. A public cloud model alone would not include the hospital's own data center as a first-class part of the design.

Question 12Cloud Computing Concepts and Business Value

A development team wants to deploy a Java application by pushing its source code to a managed cloud runtime. The provider supplies the runtime, middleware, operating system, and patching, while the team remains responsible for application code and configuration. Which service model fits?

  1. Platform as a Service (PaaS) Correct answer.
  2. Function as a Service (FaaS)
  3. Infrastructure as a Service (IaaS)
  4. Software as a Service (SaaS)

Explanation

Platform as a Service supplies a managed runtime and middleware so developers deploy application code without patching the operating system or runtime. Pushing a full Java application and leaving patching to the provider matches that division of responsibility. IaaS would leave OS and middleware management to the team, while SaaS would deliver a finished application rather than a deployment platform.

Question 13Cloud Computing Concepts and Business Value

An application running on a 4-vCPU virtual server is moved to a 16-vCPU virtual server with the same operating system and code. Throughput improves, but the architecture is unchanged. Which scaling approach was used?

  1. Autoscaling based on CPU thresholds
  2. Sharding the database across nodes
  3. Vertical scaling by increasing server capacity Correct answer.
  4. Horizontal scaling by adding more servers

Explanation

Vertical scaling adds capacity to a single server, such as moving from 4 vCPUs to 16 vCPUs while keeping the same architecture. Horizontal scaling would add more server instances behind a load balancer. Autoscaling is a policy that can trigger either approach and is not the change described.

Question 14Cloud Computing Concepts and Business Value

An IBM Cloud invoice itemizes vCPU-hours for virtual servers, GB-months for block storage, and outbound gigabytes for data transfer. Finance allocates these charges to departments based on consumption. Which essential cloud characteristic is illustrated?

  1. Measured service Correct answer.
  2. Broad network access
  3. Rapid elasticity
  4. Resource pooling

Explanation

Measured service means the provider meters resource consumption and bills or reports it by usage, enabling chargeback and optimization. Itemized vCPU-hours, GB-months, and outbound gigabytes are direct evidence of metering. Elasticity, pooling, and network access describe how resources are provisioned and reached, not how consumption is measured.

Question 15Cloud Computing Concepts and Business Value

An object storage service advertises 99.999999999 percent durability and 99.99 percent availability. How should an architect interpret these two figures?

  1. Availability is the probability an object is never lost; durability is the average retrieval latency for that object.
  2. Durability is the probability an object is not lost or corrupted; availability is the fraction of time the service is usable. Correct answer.
  3. Durability is the fraction of time the service is usable; availability is the number of replicas kept in each region.
  4. Durability and availability are the same property reported over different measurement windows and regions.

Explanation

Durability measures the probability that stored data is not lost or corrupted over time, often expressed as nines. Availability measures the proportion of time the service is operational and responding. A service can be highly durable yet briefly unavailable, and high availability does not by itself prevent data loss.

Question 16Cloud Computing Concepts and Business Value

A public cloud provider runs workloads from many customers on shared physical hosts while using logical isolation so one tenant cannot read another tenant's memory or storage. Which characteristic does this describe, and what does it require?

  1. Broad network access, which requires every workload to be reachable from the public internet
  2. Rapid elasticity, which requires workloads to scale automatically during demand spikes
  3. Resource pooling, which requires isolation controls such as hypervisors and network segmentation Correct answer.
  4. Dedicated hosting, which requires each tenant to receive its own physical server

Explanation

Resource pooling lets a provider serve multiple customers from shared physical capacity while assigning and reassigning resources dynamically. Because tenants share hardware, the provider must enforce isolation through hypervisors, virtual networks, and access controls. Rapid elasticity and broad network access are separate characteristics.

Question 17Cloud Computing Concepts and Business Value

A production database runs continuously for three years with predictable utilization, while a test environment runs only during business hours. Which purchasing approach usually gives the best balance of cost and flexibility on IBM Cloud?

  1. Purchase physical servers for production and keep the test environment in the public cloud
  2. Commit or reserve capacity for production and use hourly or on-demand pricing for test Correct answer.
  3. Use on-demand pricing for both environments to avoid any commitment
  4. Reserve capacity for the test environment and use on-demand pricing for production

Explanation

Steady, predictable workloads justify a one- or three-year commitment because the discount is earned over a known usage baseline. Variable test workloads should stay on hourly or on-demand pricing so capacity can be released when unused. Mixing committed and on-demand pricing aligns cost with utilization patterns.

Question 18Cloud Computing Concepts and Business Value

A customer deploys an application across three availability zones and signs an SLA that provides service credits for downtime. What is the most accurate interpretation of the SLA?

  1. The SLA transfers all operational risk for outages to the provider
  2. The SLA replaces the need for the customer's own disaster recovery testing
  3. The SLA guarantees that the application will never experience downtime
  4. The SLA defines provider targets and credits; resilience remains the customer's design responsibility Correct answer.

Explanation

A service-level agreement documents the provider's measurable commitments and the remedies, typically service credits, when those commitments are missed. It does not prevent outages, absorb all customer risk, or replace the customer's own resilience and recovery planning. Architects use SLAs as inputs to design decisions, not as a substitute for them.

Question 19Cloud Computing Concepts and Business Value

An application is deployed across three availability zones within a single IBM Cloud region, with synchronous replication between zones. Which failure is this design primarily intended to survive, and which failure remains outside its scope?

  1. Logical data corruption is covered; the loss of a single availability zone is not covered.
  2. Zone failure is covered; a region-wide outage or logical data corruption is not automatically covered. Correct answer.
  3. All failures are covered because three zones provide synchronous replication.
  4. Region-wide outage is covered; the loss of a single availability zone is not covered.

Explanation

Spreading a workload across three availability zones protects against the loss of one zone within the region. Synchronous replication also copies logical corruption to every zone, so it does not protect against bad writes or deletion. Surviving a full regional outage requires a separate region and an appropriate replication or backup strategy.

Question 20Cloud Computing Concepts and Business Value

An enterprise wants to keep the ability to move a containerized workload between IBM Cloud and another provider without rewriting business logic. Which design choice best supports that goal?

  1. Use OCI containers, externalize configuration, and isolate provider services behind standard interfaces Correct answer.
  2. Run the workload only on bare metal servers in one data center with manually installed software
  3. Use provider-specific event triggers and proprietary database APIs throughout the application
  4. Adopt a single vendor's serverless platform and depend on its proprietary runtime features

Explanation

Portability improves when the unit of deployment uses an open standard, configuration is externalized, and provider-specific services are hidden behind stable interfaces. OCI containers are widely supported across Kubernetes platforms, and environment-based configuration avoids baked-in host assumptions. Proprietary APIs and runtime features increase switching costs even when the code itself is containerized.

Question 21Cloud Computing Concepts and Business Value

A mid-sized firm wants employees to use hosted email, calendars, and document collaboration without installing or patching any software on company servers. Which cloud service model delivers this capability?

  1. Platform as a Service, because the firm deploys its own application code
  2. Infrastructure as a Service, because the firm still controls the operating system
  3. Container as a Service, because the firm manages container images
  4. Software as a Service, where the provider runs the finished application Correct answer.

Explanation

Software as a Service provides a complete, provider-managed application that users access over the internet. The customer does not manage servers, the runtime, or application patches, which matches the hosted email and collaboration requirement.

Question 22Cloud Computing Concepts and Business Value

An online ticket seller sees demand surge when tickets are released and fall sharply afterward. The cloud environment adds and removes capacity automatically to match demand. Which cloud characteristic does this describe?

  1. Fault tolerance
  2. Multi-tenancy
  3. Data durability
  4. Rapid elasticity Correct answer.

Explanation

Rapid elasticity is the ability to provision and release resources automatically as demand rises and falls, so capacity appears to match load closely. It is a defining measured-service characteristic of public cloud.

Question 23Cloud Computing Concepts and Business Value

In IBM Cloud, what best describes an availability zone within a region?

  1. A logical partition of a single physical server used for multi-tenant isolation
  2. A separate geographic market that has its own pricing catalog and local support team
  3. An isolated facility with independent power, cooling, and network in a region Correct answer.
  4. A billing unit that aggregates vCPU-hours across all resources in an account

Explanation

An availability zone is an isolated location within a region with independent power, cooling, and network connectivity. Distributing resources across zones reduces the chance that a single facility failure takes down the workload.

Question 24Cloud Computing Concepts and Business Value

A media company stores 80 TB of video in an IBM Cloud Object Storage bucket and streams it to viewers on the public internet every day. Which cost should the solution architect expect to be significant in addition to storage?

  1. Inbound data transfer from the internet to the bucket
  2. CPU licensing for the object storage disks
  3. Outbound data transfer to the public internet Correct answer.
  4. Fixed monthly charges for each bucket created

Explanation

Outbound data transfer to the public internet, often called egress, is commonly metered and can become a major cost for streaming workloads. Inbound transfer is usually not charged, and object storage does not require customer CPU licensing.

Question 25Cloud Computing Concepts and Business Value

A company wants a low-cost disaster recovery posture in a second region. It keeps a replicated database running there but leaves application servers stopped until a failover is declared. Which DR pattern is this?

  1. Backup and restore
  2. Multi-site active-active
  3. Warm standby with full-scale application tier
  4. Pilot light Correct answer.

Explanation

Pilot light keeps a minimal core, such as database replication, running in the recovery region and starts the remaining application tier only during failover. This provides faster recovery than backup and restore at lower cost than warm standby or active-active.

Question 26Cloud Computing Concepts and Business Value

A web application stores user session data only in the local memory of each application instance. Users are sometimes logged out when traffic shifts between instances. Which design change best supports horizontal scaling and resilience?

  1. Store session state in an external shared cache or database Correct answer.
  2. Increase the size of each instance's local memory
  3. Disable health checks so instances are not replaced
  4. Pin each user to a specific instance for the session lifetime

Explanation

Externalizing session state makes application instances stateless, so any instance can handle any request and failed instances can be replaced without losing sessions. This is a prerequisite for reliable horizontal scaling and autoscaling.

Question 27Cloud Computing Concepts and Business Value

A developer must be able to read objects in a single Object Storage bucket named reports but must not modify or delete them. Which access approach follows least privilege?

  1. Grant the account owner role to the developer for convenience
  2. Grant an IAM policy scoped to the reports bucket with a reader role Correct answer.
  3. Create a service ID with administrator access to all buckets in the account
  4. Share the account's API key with the developer

Explanation

Least privilege means granting only the permissions needed for the specific task and resource, so a bucket-scoped reader role fits. Broader roles or shared credentials give far more access than reading reports objects.

Question 28Cloud Computing Concepts and Business Value

A research organization runs steady workloads in its own data center but occasionally needs extra compute for two weeks at a time. It wants to use IBM Cloud for overflow without permanently migrating. Which approach fits?

  1. Retire the on-premises environment and use only SaaS
  2. Rewrite all applications as serverless functions before any overflow
  3. Rehost every workload permanently to IBM Cloud
  4. Use hybrid cloud bursting to send peak workloads to IBM Cloud Correct answer.

Explanation

Cloud bursting is a hybrid pattern that keeps the baseline workload on premises and sends peak demand to a public cloud for a limited time. It avoids a full migration while providing temporary additional capacity.

Question 29Cloud Computing Concepts and Business Value

A European bank must keep all customer account data physically within the European Union while using IBM Cloud managed services. Which design decision most directly satisfies this requirement?

  1. Deploy in an EU region and keep replication and backups in EU locations Correct answer.
  2. Use a content delivery network to cache account data near customers
  3. Replicate the data to every region so it is always available
  4. Encrypt the data with customer-managed keys stored in the bank's home country

Explanation

Data residency is satisfied by controlling where data is stored and copied; deploying in an EU region and keeping replication and backups in EU locations does that. Encryption and caching do not change the physical location of stored data.

Question 30Cloud Computing Concepts and Business Value

An application uses IBM Cloud Object Storage and may later move to another provider. Which practice most reduces switching cost while preserving the application's use of object storage?

  1. Keep all objects in a single bucket with no naming convention
  2. Store data in a proprietary archive tier that only one provider supports
  3. Embed the provider's SDK throughout the application and hard-code endpoint URLs
  4. Use the S3-compatible API and avoid provider-specific extensions Correct answer.

Explanation

Using a widely supported S3-compatible API limits proprietary coupling, so another provider can often be substituted with less application change. Avoiding provider-specific extensions further reduces switching cost.

Question 31Cloud Computing Concepts and Business Value

A team provisions virtual servers, block storage, and virtual networking in IBM Cloud, then installs and patches its own operating system, middleware, and application. Which cloud service model is the team using?

  1. Function as a Service (FaaS)
  2. Platform as a Service (PaaS)
  3. Software as a Service (SaaS)
  4. Infrastructure as a Service (IaaS) Correct answer.

Explanation

IaaS supplies compute, storage, and network resources while the customer remains responsible for the operating system, middleware, and application. PaaS would have the provider manage the runtime and much of the operating system, while SaaS delivers a finished application and FaaS runs event-driven functions without server management.

Question 32Cloud Computing Concepts and Business Value

A CFO compares buying servers for a three-year on-premises deployment with using IBM Cloud pay-as-you-go services. Which business-value characteristic is most directly demonstrated by the cloud option?

  1. It guarantees a lower price per compute hour than any owned server.
  2. It eliminates all security and compliance obligations for the workload.
  3. It removes the need to forecast capacity or manage costs.
  4. It converts large upfront capital spending into variable operating expenses. Correct answer.

Explanation

Pay-as-you-go cloud services shift spending from capital expenditure to operating expenditure and align cost with actual consumption. The cloud provider does not guarantee the lowest unit price, remove the need for capacity planning and cost management, or transfer the customer's compliance obligations.

Question 33Cloud Computing Concepts and Business Value

An online retailer's order service automatically adds instances within minutes during a flash sale and removes them when traffic falls. Which cloud characteristic does this behavior best illustrate?

  1. Elasticity Correct answer.
  2. Vertical scaling
  3. High availability
  4. Disaster recovery

Explanation

Elasticity is the ability to dynamically match resources to demand by scaling out and scaling in as workload changes. Vertical scaling changes the size of a single server, while high availability and disaster recovery address failures and continuity rather than demand-driven capacity adjustment.

Question 34Cloud Computing Concepts and Business Value

A company moves an application unchanged from its data center to IBM Cloud virtual servers and keeps the same manual patching, fixed capacity, and ticket-driven provisioning. Why is this lift-and-shift approach least likely to deliver the full business value of cloud?

  1. It cannot run on virtualized infrastructure unless the application is fully rewritten.
  2. It automatically turns the application into a managed SaaS offering.
  3. It removes all patching duties because the servers are virtualized.
  4. It keeps static operations, missing elasticity, automation, and usage-based savings. Correct answer.

Explanation

Lift-and-shift can move a workload to virtual servers, but much of the business value of cloud comes from changing operating practices to use elasticity, automation, and consumption-based cost. Without those changes, the organization mostly relocates the same fixed-capacity and manual processes. A virtual machine deployment does not require a rewrite, become SaaS, or eliminate guest operating system patching.

Question 35IBM Cloud Platform Architecture and Compute Options

In IBM Cloud, what is the primary purpose of deploying resources across a multizone region?

  1. To restrict all traffic to a single network zone for compliance.
  2. To provide a single dedicated host for every virtual server instance.
  3. To distribute workloads across independent data centers for higher availability. Correct answer.
  4. To reduce billing costs by sharing one physical server among tenants.

Explanation

A multizone region contains multiple physically separate zones within a metropolitan area, so you can place resources to survive a single-zone failure. Availability and latency are the goals; tenancy, network restriction, and dedicated hosts are separate design choices.

Question 36IBM Cloud Platform Architecture and Compute Options

Which IBM Cloud construct scopes resources for IAM policies and usage tracking within an account?

  1. An access group, which collects identities and applies IAM policies to them.
  2. A VLAN, which isolates broadcast traffic in classic infrastructure.
  3. A resource group, which scopes resources for IAM and billing. Correct answer.
  4. A security group, which filters traffic for virtual server instances.

Explanation

A resource group is a logical container used to organize account resources and is a scoping element in IAM policies and usage reporting. Access groups collect identities, security groups filter network traffic, and VLANs are classic network constructs.

Question 37IBM Cloud Platform Architecture and Compute Options

Which IBM Cloud compute option provides a dedicated, non-virtualized physical server allocated to a single tenant?

  1. A VPC virtual server instance with a dedicated host profile.
  2. A Kubernetes worker node in a managed cluster.
  3. A Cloud Foundry application instance in a shared cell.
  4. A bare metal server in IBM Cloud classic infrastructure. Correct answer.

Explanation

Bare metal servers are physical servers allocated to one tenant, with no hypervisor layer between the operating system and the hardware. Virtual server instances, Kubernetes workers, and Cloud Foundry cells are virtualized or shared compute.

Question 38IBM Cloud Platform Architecture and Compute Options

In an IBM Cloud VPC, how does a network access control list differ from a security group?

  1. An ACL is stateless and subnet-scoped; a security group is stateful and instance-scoped. Correct answer.
  2. An ACL applies only to classic VLANs, while a security group applies only to VPC subnets.
  3. An ACL is stateful and instance-scoped, while a security group is stateless and subnet-scoped.
  4. An ACL is enforced at the load balancer, while a security group is enforced at the subnet.

Explanation

VPC ACLs are stateless rules evaluated at the subnet boundary, so return traffic must be explicitly permitted. Security groups are stateful and attached to individual virtual server instances, automatically allowing response traffic for permitted connections.

Question 39IBM Cloud Platform Architecture and Compute Options

Why would an administrator configure multiple worker pools in an IBM Cloud Kubernetes Service cluster?

  1. To give different application teams worker nodes with different machine profiles and autoscaling policies. Correct answer.
  2. To replace the need for a load balancer by exposing pods directly on each node.
  3. To ensure every pod runs on a dedicated physical host without virtualization.
  4. To assign a separate Kubernetes API server to each namespace in the cluster.

Explanation

A worker pool groups worker nodes of the same flavor and profile, and multiple pools let you mix compute shapes and scale each group independently. The Kubernetes API server is managed by the service and is not allocated per namespace.

Question 40IBM Cloud Platform Architecture and Compute Options

What is the role of an instance group in IBM Cloud VPC?

  1. It provides a managed Kubernetes control plane for container workloads.
  2. It connects a VPC to an on-premises network through a site-to-site VPN.
  3. It maintains virtual server instances from a template and scales them with policies. Correct answer.
  4. It caches DNS records and protects applications from DDoS attacks.

Explanation

VPC instance groups use an instance template and scaling policies to keep a desired number of identical virtual servers available. They support autoscaling for stateless tiers but are not a Kubernetes control plane or a connectivity service.

Question 41IBM Cloud Platform Architecture and Compute Options

A team wants to deploy a Java application without managing servers, clusters, or containers and wants the platform to handle routing and scaling. Which IBM Cloud compute option best fits?

  1. IBM Cloud Kubernetes Service with a custom operator for deployment.
  2. IBM Cloud Bare Metal Servers with a manually installed application server.
  3. IBM Cloud Foundry using a buildpack. Correct answer.
  4. IBM Cloud VPC virtual server instances in an instance group.

Explanation

Cloud Foundry uses buildpacks to package and run applications while the platform manages routing, health checks, and scaling. It abstracts away servers and clusters, which matches the team's stated preference for minimal operational management.

Question 42IBM Cloud Platform Architecture and Compute Options

Which IBM Cloud service connects multiple VPCs and classic infrastructure networks through a single global routing domain?

  1. IBM Cloud Internet Services
  2. IBM Cloud Transit Gateway Correct answer.
  3. IBM Cloud Direct Link
  4. IBM Cloud VPC VPN gateway

Explanation

Transit Gateway attaches VPCs, classic infrastructure networks, and Direct Link connections into one routing domain, simplifying many-to-many connectivity. Direct Link provides dedicated on-premises connectivity, VPN gateways create encrypted tunnels, and Internet Services provides edge networking and security.

Question 43IBM Cloud Platform Architecture and Compute Options

A company needs private, predictable connectivity from its on-premises data center to an IBM Cloud VPC, with traffic never traversing the public internet. Which combination should it use?

  1. A VPC VPN gateway with policy-based routing over the public internet without IPsec.
  2. A public floating IP on a virtual server plus a security group allowing the data center range.
  3. IBM Cloud Internet Services with global load balancing to steer traffic to the VPC.
  4. IBM Cloud Direct Link plus Transit Gateway to route private traffic to the VPC. Correct answer.

Explanation

Direct Link provides dedicated private connectivity between the customer network and IBM Cloud, and Transit Gateway attaches the VPC and Direct Link connection for routing. Public floating IPs and Internet Services use public paths, while a VPC VPN gateway normally uses encrypted tunnels over the internet.

Question 44IBM Cloud Platform Architecture and Compute Options

A regulated customer wants a managed Kubernetes platform with built-in developer tooling and stricter default security controls, and it wants the option to run the same operational model in its own data center. Which IBM Cloud offering aligns best?

  1. IBM Cloud Foundry, because buildpacks provide the strictest container isolation by default.
  2. IBM Cloud Kubernetes Service, because it provides only a managed control plane and no integrated developer tooling.
  3. IBM Cloud VPC instance groups, because they provide a Kubernetes-compatible autoscaling API.
  4. Red Hat OpenShift on IBM Cloud, which adds OpenShift tooling and can extend on-premises with Satellite. Correct answer.

Explanation

Red Hat OpenShift on IBM Cloud is a managed OpenShift service that includes developer and CI/CD tooling plus security context constraints. IBM Cloud Satellite can extend that managed service pattern to customer data centers, which addresses the on-premises requirement.

Question 45IBM Cloud Platform Architecture and Compute Options

In IBM Cloud VPC, what is the scope of a subnet?

  1. A single zone within a region Correct answer.
  2. All VPCs in the account
  3. The entire multizone region
  4. The account's resource group

Explanation

A VPC spans a region, but each subnet is created in one specific zone and uses an address prefix from that zone. This zonal scope is what lets a VPC place workloads in different zones for availability while sharing the regional VPC network.

Question 46IBM Cloud Platform Architecture and Compute Options

A virtual server instance in an IBM Cloud VPC has no floating IP address but must download software updates from the internet. What should be attached to the subnet to allow outbound internet access?

  1. A public gateway Correct answer.
  2. A security group with an allow-all rule
  3. A site-to-site VPN gateway
  4. A network ACL rule for all outbound traffic

Explanation

A public gateway provides outbound connectivity from a VPC subnet to the internet, and it performs NAT so instances can use their private addresses. Without a public gateway or a floating IP, an instance has no route to public destinations.

Question 47IBM Cloud Platform Architecture and Compute Options

In IBM Cloud Kubernetes Service, which component is fully managed by IBM Cloud and runs the Kubernetes API server and etcd?

  1. The Ingress application load balancer
  2. The persistent volume claim
  3. The worker nodes in each worker pool
  4. The cluster control plane Correct answer.

Explanation

IBM Cloud manages the cluster control plane, including the Kubernetes API server and etcd, so the customer does not patch or operate those components. Customers are responsible for the worker nodes and the workloads they run.

Question 48IBM Cloud Platform Architecture and Compute Options

An application team needs to distribute HTTPS requests across virtual server instances in two subnets of one VPC and terminate TLS at the load balancer. Which IBM Cloud VPC resource should they deploy?

  1. An application load balancer Correct answer.
  2. A transit gateway
  3. A network access control list
  4. A public gateway

Explanation

The VPC application load balancer distributes HTTP and HTTPS traffic across backend instances and can terminate TLS using a certificate. It can target instances in multiple subnets of the same VPC, making it the right choice here.

Question 49IBM Cloud Platform Architecture and Compute Options

Which IBM Cloud VPC resource provides a physically isolated physical host on which a customer can provision multiple virtual server instances with controlled placement?

  1. A dedicated host Correct answer.
  2. A placement group
  3. An instance group
  4. A bare metal server

Explanation

A dedicated host is a single-tenant physical server that hosts virtual server instances, giving the customer control over placement and host-level isolation. It differs from a bare metal server, which is itself the compute instance rather than a host for multiple VSIs.

Question 50IBM Cloud Platform Architecture and Compute Options

A workload in a VPC must access an IBM Cloud service over the private network without using the public internet. Which VPC resource provides a private IP endpoint for the service?

  1. A site-to-site VPN gateway
  2. An endpoint gateway Correct answer.
  3. A floating IP address
  4. A public gateway

Explanation

An endpoint gateway creates private IP addresses in a VPC subnet for supported IBM Cloud services, so traffic stays on the IBM Cloud private network. It removes the need for public internet egress or a VPN to reach those services.