I85PREP
⌘ K
Akamai Cloud Computing Foundations study materials
Akamai Cloud Computing Foundations study resources

Akamai Cloud Computing Foundations practice questions

One question at a time. Build your Akamai Cloud Computing Foundations understanding.

First 50 of 199 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 Fundamentals

Which cloud service model delivers a managed environment where developers deploy code while the provider handles the operating system and runtime?

  1. Infrastructure as a Service
  2. Desktop as a Service
  3. Software as a Service
  4. Platform as a Service Correct answer.

Explanation

Platform as a Service supplies a managed runtime and middleware so developers focus on application code. The provider patches and scales the underlying platform rather than exposing the operating system for customer administration.

Question 2Cloud Computing Fundamentals

Which service model typically gives a customer the most direct control over virtual machines, storage, and network configuration?

  1. Infrastructure as a Service Correct answer.
  2. Software as a Service
  3. Platform as a Service
  4. Function as a Service

Explanation

IaaS provides virtualized compute, storage, and networking resources that the customer configures and secures. The customer generally manages the guest operating system and middleware on top of that infrastructure.

Question 3Cloud Computing Fundamentals

A company uses cloud resources owned by a third-party provider and shared with other organizations over the public internet. Which deployment model is this?

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

Explanation

A public cloud is owned and operated by a provider and serves multiple customers over a network. Resources are pooled across tenants even though each customer receives logical isolation.

Question 4Cloud Computing Fundamentals

Which characteristic of cloud computing lets a consumer provision resources such as servers automatically without human interaction from the provider?

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

Explanation

On-demand self-service allows a consumer to obtain computing capabilities through automation without requiring human interaction with the provider. Resource pooling, measured service, and elasticity describe other aspects of the cloud service model.

Question 5Cloud Computing Fundamentals

What does elasticity in cloud computing describe?

  1. Reserving a fixed amount of capacity for a long term
  2. Encrypting data at rest across storage services
  3. Matching resource capacity to demand as it changes Correct answer.
  4. Moving workloads between providers to reduce cost

Explanation

Elasticity is the ability to acquire and release resources so that capacity tracks changing demand. It supports scaling out during peaks and scaling in when demand falls.

Question 6Cloud Computing Fundamentals

Which billing approach is most characteristic of public cloud consumption?

  1. Usage-based charges for consumed resources Correct answer.
  2. A one-time perpetual license fee
  3. A fixed annual fee regardless of usage
  4. A per-employee fee for support tickets

Explanation

Public cloud billing commonly meters consumed resources such as compute hours, storage, and data transfer. Customers pay for what they use rather than buying a perpetual license or paying a flat annual fee.

Question 7Cloud Computing Fundamentals

What is the primary role of a hypervisor in cloud infrastructure?

  1. To store objects durably across multiple disks
  2. To route network traffic between regions
  3. To compile application code into containers
  4. To run virtual machines on physical hosts Correct answer.

Explanation

A hypervisor creates and manages virtual machines by partitioning physical host resources among guests. It isolates guest operating systems from the underlying hardware and from one another.

Question 8Cloud Computing Fundamentals

What is an availability zone in a cloud region?

  1. A software-defined firewall for a virtual network
  2. A global edge site used only for content caching
  3. A billing boundary that groups subscriptions
  4. An isolated site with independent power and network in a region Correct answer.

Explanation

An availability zone is an isolated location within a region with independent power, cooling, and networking. Applications can use multiple zones to reduce the risk of a single-site failure.

Question 9Cloud Computing Fundamentals

Adding more server instances to share an application workload is an example of which scaling method?

  1. Vertical scaling
  2. Horizontal scaling Correct answer.
  3. Manual provisioning
  4. Diagonal scaling

Explanation

Horizontal scaling, or scaling out, adds more instances so that workload is distributed across them. Vertical scaling instead changes the size or power of a single instance.

Question 10Cloud Computing Fundamentals

Under the shared responsibility model for a virtual machine in a public cloud, which task generally belongs to the customer?

  1. Securing the hypervisor and physical hosts
  2. Patching the guest OS and applications Correct answer.
  3. Maintaining the physical data center cooling systems
  4. Managing the provider's global network backbone

Explanation

For IaaS virtual machines, the provider secures the physical facilities and hypervisor while the customer secures the guest operating system, applications, and data. The boundary depends on the service model, but guest patching is normally a customer task.

Question 11Cloud Computing Fundamentals

Which workload is best suited to object storage rather than block storage?

  1. A shared file system for active directory roaming profiles
  2. A database requiring low-latency random writes to a volume
  3. Storing large backups and media files accessed by URL Correct answer.
  4. Hosting the boot volume for a virtual machine

Explanation

Object storage uses a flat namespace and HTTP-accessible objects with metadata, making it well suited to backups, media, and unstructured content. Block storage is better for low-latency random I/O such as databases and boot volumes.

Question 12Cloud Computing Fundamentals

Which statement describes a defining characteristic of serverless compute?

  1. Billing is based on reserved capacity rather than requests
  2. The customer retains root access to the underlying host
  3. The provider manages servers and scales functions on demand Correct answer.
  4. The customer must pre-provision a fixed number of hosts

Explanation

Serverless compute hides server management and scales execution in response to events or requests. Billing is commonly based on invocations and execution duration rather than reserved hosts.

Question 13Cloud Computing Fundamentals

What is the main purpose of a load balancer in a cloud environment?

  1. To archive log data for long-term retention
  2. To encrypt traffic between two virtual networks
  3. To spread incoming requests across backends Correct answer.
  4. To assign public IP addresses to databases

Explanation

A load balancer distributes client requests across multiple backend targets to improve availability and utilization. It can also perform health checks and remove unhealthy targets from rotation.

Question 14Cloud Computing Fundamentals

Which condition is most appropriate as a trigger for autoscaling a web tier?

  1. The color scheme of the application dashboard
  2. The domain name registration expiry date
  3. The number of administrator accounts
  4. CPU utilization exceeding a set threshold Correct answer.

Explanation

Autoscaling policies commonly react to metrics such as CPU, memory, request rate, or queue depth. A CPU threshold reflects workload pressure and can trigger adding or removing instances.

Question 15Cloud Computing Fundamentals

Compared with virtual machines, containers typically share which component?

  1. A unique hypervisor per application
  2. A dedicated physical network interface
  3. A separate guest operating system for each instance
  4. The host operating system kernel Correct answer.

Explanation

Containers share the host operating system kernel and isolate processes using features such as namespaces and control groups. Virtual machines each include a full guest operating system and rely on a hypervisor.

Question 16Cloud Computing Fundamentals

According to the CAP theorem, what must a distributed system trade off during a network partition?

  1. Storage and compute
  2. Latency and durability
  3. Encryption and authentication
  4. Consistency and availability Correct answer.

Explanation

CAP states that a distributed system cannot simultaneously guarantee consistency and availability during a network partition. Designers choose how to behave when partitions occur, such as favoring one property over the other.

Question 17Cloud Computing Fundamentals

Which scenario best describes a hybrid cloud deployment?

  1. A private data center is retired completely
  2. On-premises systems integrate with public cloud services Correct answer.
  3. Multiple tenants share one provider's database cluster
  4. All workloads run in a single provider's public region

Explanation

A hybrid cloud combines distinct environments, such as on-premises infrastructure and public cloud services, with integration between them. Data and workloads can move or interact across the environments.

Question 18Cloud Computing Fundamentals

What primary benefit does a content delivery network provide for static web assets?

  1. It provides unlimited compute for background jobs
  2. It caches content closer to users to reduce latency Correct answer.
  3. It encrypts databases at rest automatically
  4. It replaces the need for load balancers in all cases

Explanation

A CDN caches content at edge locations near users, reducing the distance requests travel and improving response times. It can also reduce origin load and absorb some traffic spikes.

Question 19Cloud Computing Fundamentals

In disaster recovery planning, what does recovery time objective (RTO) measure?

  1. The cost of replicated storage per region
  2. Target time to restore service after an outage Correct answer.
  3. The maximum acceptable amount of data loss
  4. The number of backups retained each year

Explanation

RTO defines the target duration for restoring a service after a disruption. It is distinct from recovery point objective, which describes acceptable data loss.

Question 20Cloud Computing Fundamentals

What does multi-tenancy mean in a public cloud platform?

  1. Workloads are limited to a single availability zone
  2. Multiple customers share infrastructure with logical isolation Correct answer.
  3. A tenant can access another tenant's billing records
  4. Each customer receives a dedicated physical data center

Explanation

Multi-tenancy pools physical resources among customers while isolating their workloads and data logically. The provider manages the sharing so tenants do not interfere with one another.

Question 21Cloud Computing Fundamentals

An application must survive the loss of a single data center without downtime. Which architecture best supports this requirement?

  1. Spread instances across zones behind a load balancer Correct answer.
  2. Store application state on each instance's local disk
  3. Deploy all instances in one zone with nightly backups
  4. Use a single large virtual machine with a public IP

Explanation

Distributing instances across availability zones removes a single data center as a single point of failure. A load balancer routes traffic to healthy instances in the remaining zones when one zone fails.

Question 22Cloud Computing Fundamentals

Which practice best reduces vendor lock-in when adopting cloud services?

  1. Store all data in a single provider's managed database
  2. Use open standards and portable interfaces where possible Correct answer.
  3. Avoid automation so migrations remain manual
  4. Use proprietary APIs for every service to maximize features

Explanation

Portable interfaces and open standards make it easier to move workloads or integrate alternative providers. Lock-in increases when critical data and logic depend on proprietary services that are difficult to replicate elsewhere.

Question 23Cloud Computing Fundamentals

What is a key benefit of managing cloud infrastructure with infrastructure as code?

  1. It guarantees zero cost for all deployed resources
  2. It removes the need to test infrastructure changes
  3. It eliminates the need for any access controls
  4. It makes environments reproducible and versioned Correct answer.

Explanation

Infrastructure as code defines resources in version-controlled files, so environments can be recreated consistently. Changes can be reviewed and tested in the same way as application code.

Question 24Cloud Computing Fundamentals

When using a managed database service, which responsibility shifts to the cloud provider compared with self-managed database software on a virtual machine?

  1. Writing and tuning application queries
  2. Controlling which users can read application tables
  3. Applying database engine patches and backups Correct answer.
  4. Defining the application's schema and indexes

Explanation

A managed database service typically handles engine patching, backups, and routine platform maintenance. The customer remains responsible for schema design, query performance, and data access controls.

Question 25Cloud Computing Fundamentals

Which design choice is most consistent with building a cloud-native application?

  1. Saving session state on the local disk of each instance
  2. Using stateless services and managed backing services Correct answer.
  3. Treating servers as long-lived pets that are manually repaired
  4. Deploying one large monolithic process for all features

Explanation

Cloud-native applications favor stateless, replaceable service instances and managed backing services for persistence. This supports automated scaling, rolling deployments, and recovery from instance failure.

Question 26Cloud Computing Fundamentals

Which statement best describes rapid elasticity in cloud computing?

  1. Resources are purchased in fixed annual blocks for predictable pricing.
  2. Resources are limited to a single physical data center.
  3. Resources are shared only among tenants in the same department.
  4. Resources are added or removed automatically to match demand closely. Correct answer.

Explanation

Rapid elasticity is the capability to provision and release resources automatically as demand changes, so capacity closely tracks actual load. It is a defining characteristic of cloud computing and does not imply fixed annual purchasing or single-site limits.

Question 27Cloud Computing Fundamentals

What does on-demand self-service allow a cloud consumer to do?

  1. The provider schedules capacity only during business hours.
  2. The provider manually approves each resource request before provisioning.
  3. A consumer provisions resources without requiring provider staff interaction. Correct answer.
  4. Consumers must sign a new contract for every virtual machine.

Explanation

On-demand self-service lets a consumer provision computing capabilities, such as server time or storage, without human interaction with the provider. This automation is what enables rapid, unassisted resource acquisition.

Question 28Cloud Computing Fundamentals

Which characteristic distinguishes a public cloud from a private cloud?

  1. The infrastructure is owned and operated by the customer's IT staff.
  2. Resources are reserved exclusively for a single government agency.
  3. A third-party provider owns the infrastructure and delivers services over a network. Correct answer.
  4. The cloud is accessible only from the customer's private network.

Explanation

In a public cloud, a third-party provider owns and operates the infrastructure and makes services available to multiple customers over a network. Customers consume resources rather than owning the underlying hardware.

Question 29Cloud Computing Fundamentals

Which responsibility pattern is typical of Software as a Service?

  1. The customer manages the application code and the provider manages only the hardware.
  2. The provider supplies only raw compute and storage, leaving the platform to the customer.
  3. The provider manages the application and infrastructure, and users access it through a client. Correct answer.
  4. The customer installs and patches the application on provider-managed virtual machines.

Explanation

Software as a Service delivers a complete application that the provider operates, including the underlying platform and infrastructure. Users access it through a client such as a browser, without managing servers or application code.

Question 30Cloud Computing Fundamentals

A team deploys code to a managed runtime and no longer patches the operating system or runtime software. Which service model does this describe?

  1. The provider manages the runtime; the team manages application code. Correct answer.
  2. The provider writes and maintains all application business logic.
  3. The team patches the operating system and the runtime environment.
  4. The team manages physical servers and the hypervisor layer.

Explanation

With Platform as a Service, the provider manages the runtime, middleware, and operating system, while the customer remains responsible for the application code and data. This division reduces operational burden without transferring business logic ownership.

Question 31Cloud Computing Fundamentals

A database is CPU-bound, and the administrator moves it to a larger instance with more vCPUs and memory. Which scaling approach is this?

  1. Add more web servers behind the load balancer.
  2. Distribute the database across multiple shards.
  3. Add a read replica in another region.
  4. Move the database to a larger instance with more vCPUs. Correct answer.

Explanation

Vertical scaling increases the capacity of a single instance, for example by moving a database to a larger size with more vCPUs and memory. Adding instances or shards would be horizontal scaling instead.

Question 32Cloud Computing Fundamentals

A platform can grow to serve millions of users, but capacity must be planned manually weeks in advance. Which statement best describes it?

  1. It is elastic but not scalable.
  2. It is neither scalable nor elastic.
  3. It is both highly elastic and highly scalable.
  4. It is scalable but not elastic. Correct answer.

Explanation

Scalability is the ability to handle growth, while elasticity is the ability to match capacity to demand automatically and quickly. A system that can grow but needs weeks of manual planning is scalable yet not elastic.

Question 33Cloud Computing Fundamentals

Which deployment model is shared by several organizations with common requirements and governed by a shared policy?

  1. Owned by one company and leased exclusively to one tenant.
  2. Dedicated to a single enterprise and hosted on its premises.
  3. Open to any customer and billed on a pay-as-you-go basis.
  4. Shared by organizations with common requirements under a common policy. Correct answer.

Explanation

A community cloud is shared by several organizations that have common concerns, such as compliance or mission requirements, and is governed by a shared policy. It may be managed by the organizations or by a third party.

Question 34Cloud Computing Fundamentals

A retailer runs steady traffic in its private data center and sends overflow to a public cloud during seasonal peaks. Which term best describes this pattern?

  1. Vertical scaling, resizing the private database to absorb the peak.
  2. Disaster recovery failover, moving all workloads after a site outage.
  3. Multi-tenancy, sharing one application instance among many customers.
  4. Cloud bursting, using public capacity when private capacity is exhausted. Correct answer.

Explanation

Cloud bursting uses a private environment for normal load and adds public cloud capacity when demand exceeds private capacity. It is a hybrid pattern for handling temporary peaks without permanently expanding the private data center.

Question 35Cloud Computing Fundamentals

Which design choice makes it easiest to add or replace application servers without disrupting user sessions?

  1. Replicate the session file across every server nightly.
  2. Pin each user to one server for the duration of their session.
  3. Persist session data on the local disk of each application server.
  4. Keep session state in a shared external store accessible to all instances. Correct answer.

Explanation

A stateless application tier keeps session state in a shared external store, so any instance can handle any request. This allows requests to be distributed freely and lets instances be added or replaced without session affinity.

Question 36Cloud Computing Fundamentals

A company must keep customer data physically within Germany. Which cloud design choice best satisfies this requirement?

  1. Enable encryption at rest for all storage volumes.
  2. Replicate the data to three continents for durability.
  3. Choose a cloud region inside Germany for all data storage. Correct answer.
  4. Use a global content delivery network for faster access.

Explanation

Data residency requirements are met by choosing a region, and its underlying facilities, inside the required legal jurisdiction. Encryption and replication do not change where the data is physically stored.

Question 37Cloud Computing Fundamentals

A provider offers a 99.9% monthly availability SLA. Approximately how much downtime does that permit in a 30-day month?

  1. About 7 minutes
  2. About 43 minutes Correct answer.
  3. About 4 hours
  4. About 22 minutes

Explanation

A 99.9% monthly SLA permits 0.1% unavailability. In a 30-day month of 43,200 minutes, 0.1% is about 43.2 minutes of allowed downtime.

Question 38Cloud Computing Fundamentals

Which change best describes the financial shift from capital expenditure to operational expenditure in cloud adoption?

  1. Upfront hardware buys become recurring usage-based fees. Correct answer.
  2. Monthly operational costs are converted into large capital purchases.
  3. Hardware refresh cycles become the customer's responsibility.
  4. Software licensing costs are eliminated entirely by cloud providers.

Explanation

Cloud computing shifts spending from capital expenditure to operational expenditure: organizations pay recurring usage-based fees instead of making large upfront hardware purchases. This can improve cash flow and align cost with consumption.

Question 39Cloud Computing Fundamentals

What does a durability rating of eleven nines for object storage indicate?

  1. Data is copied to exactly eleven geographic regions.
  2. Stored objects are highly unlikely to be lost or corrupted over time. Correct answer.
  3. Read requests will complete within 11 milliseconds.
  4. The service will be reachable 99.999999999% of the time.

Explanation

Durability describes the likelihood that stored data remains intact and is not lost or corrupted over time. Eleven nines durability indicates an extremely low annual probability of object loss; it does not describe service uptime or latency.

Question 40Cloud Computing Fundamentals

A web tier writes jobs directly to a worker database, and traffic spikes cause worker failures and lost jobs. Which change best decouples the tiers?

  1. Increase the worker database instance size.
  2. Move the worker database to the same server as the web tier.
  3. Add a second web server behind the load balancer.
  4. Put a durable message queue between the tiers. Correct answer.

Explanation

A message queue decouples producers from consumers, absorbing spikes and allowing the worker tier to process jobs at its own rate. This prevents the web tier from overwhelming the workers and reduces lost work.

Question 41Compute and Storage Services

Which compute instance plan family offers the lowest-cost general-purpose option, where vCPUs may be shared with other workloads on the same host?

  1. GPU
  2. Dedicated CPU
  3. Shared CPU Correct answer.
  4. High Memory

Explanation

Shared CPU plans are the entry-level family; their vCPUs are scheduled alongside other customers' workloads, which keeps cost low but allows variable performance. Dedicated, High Memory, and GPU plans are specialized families with different pricing and guarantees.

Question 42Compute and Storage Services

A Block Storage volume is designed for which attachment model?

  1. Unlimited compute instances within the same region
  2. One compute instance per data center
  3. Up to two compute instances in the same data center
  4. Exactly one compute instance at a time Correct answer.

Explanation

A volume is attached to a single compute instance at any moment. To move it, you detach it and attach it to another instance in the same data center, but concurrent attachment to multiple instances is not supported.

Question 43Compute and Storage Services

Object Storage in Akamai Cloud Computing is accessed through which interface?

  1. iSCSI block targets
  2. An S3-compatible API Correct answer.
  3. SFTP only
  4. NFSv4 mounts

Explanation

Object Storage exposes an S3-compatible HTTP API, so S3 SDKs and command-line tools work against it. It is neither a block device nor a native POSIX filesystem, so NFS and iSCSI are not the access method.

Question 44Compute and Storage Services

What is the primary purpose of a StackScript?

  1. To automate instance deployment with a reusable script Correct answer.
  2. To monitor billing across accounts
  3. To encrypt Block Storage volumes at rest
  4. To host DNS zones for a domain

Explanation

StackScripts are scripts stored in your account that run during instance creation to install packages and configure services. They can include user-defined fields so each deployment supplies different values without rewriting the script.

Question 45Compute and Storage Services

Which workload is the best fit for a GPU compute instance?

  1. A DNS resolver with light traffic
  2. Machine learning training and rendering Correct answer.
  3. A small static brochure site
  4. Long-term cold archival of files

Explanation

GPU plans attach accelerators suited to parallel math workloads such as model training, inference, and image rendering. Lightweight web or DNS workloads do not benefit from a GPU and would waste the added cost.

Question 46Compute and Storage Services

When a compute instance is deleted, what happens to its attached Block Storage volume?

  1. The volume is deleted with the instance
  2. The volume remains and can be attached elsewhere Correct answer.
  3. The volume is moved to Object Storage
  4. The volume is converted into an image automatically

Explanation

Block Storage volumes are independent resources with their own lifecycle, so they survive instance deletion and can be reattached to another instance in the same data center. Local disks tied to the instance do not survive in the same way.

Question 47Compute and Storage Services

Which action replaces the disks of an existing compute instance with a chosen image while preserving its IP addresses?

  1. Resizing the instance
  2. Rebuilding the instance Correct answer.
  3. Cloning the instance
  4. Enabling backups

Explanation

Rebuild wipes the instance's disks and reinstalls from a selected image while keeping the instance's networking identity. Because it destroys data on those disks, backups should be taken first.

Question 48Compute and Storage Services

What does cloning a compute instance produce?

  1. A backup that can be restored later
  2. A snapshot stored in Object Storage
  3. A second instance that shares the same disks
  4. A new independent instance with copies of the source disks Correct answer.

Explanation

A clone is a new instance built from copies of the source instance's disks, so later changes on one do not affect the other. It is a separate instance with its own lifecycle and billing.

Question 49Compute and Storage Services

An application runs in the Frankfurt data center and needs a new Block Storage volume. What placement is required?

  1. Create the volume in any data center and attach it over the public internet
  2. Create the volume in the account's default data center
  3. Create the volume in the same region as your Object Storage bucket
  4. Create the volume in Frankfurt so it can attach to the instance Correct answer.

Explanation

Volumes attach only to compute instances in the same data center, so the volume must be created in Frankfurt. Cross-data-center attachment is not possible, and neither Object Storage placement nor a default data center setting changes that.

Question 50Compute and Storage Services

Which change can be made to an existing Block Storage volume's size?

  1. Increase it, but not decrease it Correct answer.
  2. Change it only by opening a support ticket
  3. Shrink it to any smaller size
  4. Change it freely in either direction

Explanation

Volumes can be enlarged to accommodate growth, but they cannot be shrunk because reducing the block device size could destroy data. Resizing is a self-service operation rather than a support-only request.