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 Fundamentals
Which characteristic of cloud computing lets a provider add or remove resources automatically to match current demand without manual intervention?
- Broad network access
- Rapid elasticity Correct answer.
- Multi-tenancy
- Measured service
Explanation
Rapid elasticity is the capability to scale resources outward and inward automatically to match demand. It focuses on dynamic provisioning rather than sharing, metering, or access.
Question 2Cloud Computing Fundamentals
A company wants to move from buying servers to paying only for the compute hours it consumes. Which cloud benefit does this describe?
- Improved disaster recovery
- Reduced capital expenditure Correct answer.
- Increased vendor lock-in
- Higher latency
Explanation
Converting up-front hardware purchases into usage-based operating costs reduces capital expenditure. This is the pay-as-you-go economic model of cloud computing.
Question 3Cloud Computing Fundamentals
Which service model provides a runtime environment where developers deploy code but do not manage the underlying operating system?
- Platform as a Service Correct answer.
- Infrastructure as a Service
- Software as a Service
- Function as a Service
Explanation
Platform as a Service supplies a managed runtime and middleware so developers focus on application code. The provider handles the OS and infrastructure layers.
Question 4Cloud Computing Fundamentals
In the shared responsibility model for IaaS, which task remains the customer's responsibility?
- Patching the guest operating system Correct answer.
- Replacing failed physical disks
- Securing the hypervisor
- Maintaining physical data center security
Explanation
Under IaaS the provider secures facilities and virtualization, while the customer secures and patches the guest OS and everything above it. Physical and hypervisor layers belong to the provider.
Question 5Cloud Computing Fundamentals
Which deployment model places workloads in a provider's data center while keeping them isolated to a single organization?
- Hybrid cloud
- Private cloud Correct answer.
- Community cloud
- Public cloud
Explanation
A private cloud is provisioned for exclusive use by a single organization, whether on or off premises. Isolation to one tenant is its defining trait.
Question 6Cloud Computing Fundamentals
Which cloud deployment model is shared by several organizations that have common compliance requirements?
- Private cloud
- Public cloud
- Edge cloud
- Community cloud Correct answer.
Explanation
A community cloud is used by a specific group of organizations with shared concerns such as compliance or mission. It is neither exclusive to one tenant nor open to the general public.
Question 7Cloud Computing Fundamentals
Which technology enables multiple isolated virtual servers to run on one physical host?
- Hypervisor Correct answer.
- Content delivery network
- Object storage gateway
- Load balancer
Explanation
A hypervisor creates and runs virtual machines, abstracting physical hardware into multiple isolated guests. It is the core enabler of server virtualization.
Question 8Cloud Computing Fundamentals
Which storage type presents block-level volumes that a virtual machine formats with its own file system?
- Archive storage
- File storage
- Object storage
- Block storage Correct answer.
Explanation
Block storage exposes raw volumes that attach to a compute instance and are formatted by the guest OS. This differs from file-level shares and object buckets.
Question 9Cloud Computing Fundamentals
Which cloud storage model is best suited for storing millions of unstructured files accessed by HTTP API?
- Local instance store
- Object storage Correct answer.
- Block storage
- SAN volume
Explanation
Object storage keeps data as objects with metadata in flat namespaces and is accessed over HTTP APIs, scaling to huge numbers of unstructured items. Block and SAN volumes are not designed for API-based object access at that scale.
Question 10Cloud Computing Fundamentals
What is the primary purpose of a content delivery network in a cloud architecture?
- Metering compute usage
- Caching content at edge locations near users Correct answer.
- Provisioning virtual machines
- Encrypting data at rest
Explanation
A CDN caches content at geographically distributed edge nodes to reduce latency and origin load. Encryption, provisioning, and metering are separate functions.
Question 11Cloud Computing Fundamentals
Which networking construct isolates resources in a private, logically defined network within a public cloud?
- Public IP address
- Domain name system
- Content delivery network
- Virtual private cloud Correct answer.
Explanation
A virtual private cloud provides an isolated logical network with its own address ranges and routing. Public IPs, CDNs, and DNS operate at other layers.
Question 12Cloud Computing Fundamentals
Which load balancing algorithm sends each new request to the server with the fewest active connections?
- IP hash
- Round robin
- Least connections Correct answer.
- Weighted random
Explanation
Least connections tracks active connection counts and routes to the least busy server. This adapts to uneven request durations better than simple rotation.
Question 13Cloud Computing Fundamentals
Which approach improves availability by distributing a workload across multiple availability zones?
- Using a single larger instance
- Disabling health checks
- Storing backups in the same zone
- Deploying redundant instances in separate zones Correct answer.
Explanation
Availability zones are isolated failure domains, so replicating across zones lets a workload survive a zone outage. A single instance or same-zone backups concentrate risk.
Question 14Cloud Computing Fundamentals
A team needs to run a batch job that can tolerate interruption and wants the lowest compute cost. Which option fits best?
- Dedicated hosts
- Spot or preemptible instances Correct answer.
- On-demand instances
- Reserved instances
Explanation
Spot or preemptible capacity is discounted in exchange for possible interruption, which suits fault-tolerant batch work. Reserved and dedicated options cost more for guaranteed capacity.
Question 15Cloud Computing Fundamentals
Which pricing model offers a discount in exchange for committing to a defined usage level over a one- or three-year term?
- Reserved capacity Correct answer.
- Spot
- On-demand
- Per-request
Explanation
Reserved capacity discounts reward a long-term commitment to a predictable amount of usage. On-demand and spot do not require such a term commitment.
Question 16Cloud Computing Fundamentals
Which concept describes automatically adding instances when CPU utilization exceeds a threshold?
- Sharding
- Caching
- Vertical scaling
- Horizontal scaling Correct answer.
Explanation
Adding more instances spreads load across machines and is horizontal (scale out) scaling. Vertical scaling instead increases the size of a single instance.
Question 17Cloud Computing Fundamentals
Which scaling approach increases the CPU and memory of an existing server rather than adding servers?
- Horizontal scaling
- Read replica addition
- Vertical scaling Correct answer.
- Auto-scaling group expansion
Explanation
Vertical scaling, or scaling up, resizes a single machine to give it more resources. It does not change the number of instances.
Question 18Cloud Computing Fundamentals
What does a service level agreement typically define between a cloud provider and a customer?
- The source code of the provider's platform
- The physical layout of the data center
- The customer's internal security policies
- Guaranteed performance and availability targets with remedies Correct answer.
Explanation
An SLA states measurable service commitments such as uptime and the credits or remedies if they are missed. It does not disclose provider internals or dictate customer policy.
Question 19Cloud Computing Fundamentals
Which practice supports the principle of least privilege in a cloud environment?
- Sharing a single administrator account
- Disabling audit logging
- Granting identities only the permissions their tasks require Correct answer.
- Storing credentials in source code
Explanation
Least privilege limits each identity to exactly the permissions needed, reducing blast radius if compromised. Shared admin accounts and embedded credentials do the opposite.
Question 20Cloud Computing Fundamentals
Which control most directly protects data at rest in cloud storage?
- TLS on the network path
- Multi-factor authentication
- Encryption using provider-managed or customer-managed keys Correct answer.
- A web application firewall
Explanation
Encryption at rest encodes stored data so it is unreadable without keys, protecting it if media is exposed. TLS, WAF, and MFA address other layers.
Question 21Cloud Computing Fundamentals
Which cloud capability lets an organization automatically rebuild infrastructure from version-controlled templates?
- Password vaulting
- Manual console provisioning
- Infrastructure as code Correct answer.
- Snapshot-only backups
Explanation
Infrastructure as code defines environments in versioned templates so they can be reproduced and audited consistently. Manual provisioning is error-prone and not reproducible.
Question 22Cloud Computing Fundamentals
A workload has strict data residency requirements and cannot leave a specific country. Which cloud consideration is most relevant?
- Container registry choice
- Load balancer algorithm
- Instance type family
- Region selection for data location Correct answer.
Explanation
Data residency is governed by where data is physically stored, so choosing an in-country region is the controlling decision. Instance type, balancing, and registries do not determine legal location.
Question 23Cloud Computing Fundamentals
Which statement correctly distinguishes fault tolerance from high availability?
- High availability guarantees zero downtime, while fault tolerance allows brief outages
- Fault tolerance and high availability are identical in every respect
- Fault tolerance continues operating through a component failure with no interruption, while high availability minimizes downtime Correct answer.
- Fault tolerance only applies to storage, while high availability only applies to compute
Explanation
Fault-tolerant designs mask failures so service continues without interruption, whereas highly available designs aim to restore service quickly with minimal downtime. The two differ in whether an interruption is permitted at all.
Question 24Cloud Computing Fundamentals
A team must choose between a managed database service and self-managed databases on virtual machines. Which factor most favors the managed service?
- Offloading patching, backups, and failover to the provider Correct answer.
- Full control over the storage engine internals
- Ability to install arbitrary kernel modules
- Direct root access to the database host
Explanation
Managed database services take over operational tasks such as patching, backups, and failover, reducing administrative burden. Self-managed options are chosen when deep host-level control is required.
Question 25Cloud Computing Fundamentals
Which combination best characterizes a hybrid cloud architecture?
- Integrating on-premises systems with public cloud services through secure connectivity Correct answer.
- Using only a private cloud with no external links
- Running all workloads exclusively in one public provider
- Sharing infrastructure among organizations with common compliance needs
Explanation
Hybrid cloud links private or on-premises environments with public cloud, typically via secure networking, so workloads can span both. Single-provider and purely private setups are not hybrid.
Question 26Cloud Computing Fundamentals
Which characteristic of cloud computing lets a customer add or remove resources in response to demand without long procurement delays?
- Rapid elasticity Correct answer.
- Manual provisioning
- Fixed capacity
- Vendor lock-in
Explanation
Rapid elasticity means resources can be acquired and released automatically or on demand as workload changes, closely matching supply to demand. The other choices describe constraints or manual processes rather than an on-demand scaling property.
Question 27Cloud Computing Fundamentals
Which cloud service model gives the customer the greatest amount of control over the operating system and runtime environment?
- Software as a Service
- Infrastructure as a Service Correct answer.
- Platform as a Service
- Function as a Service
Explanation
In IaaS the provider supplies compute, storage, and networking while the customer manages the OS, middleware, and applications. Higher-level models such as PaaS and SaaS abstract the OS away from the customer.
Question 28Cloud Computing Fundamentals
What is the primary benefit of an object storage lifecycle policy that transitions data to a colder tier after 30 days?
- It increases data durability
- It removes the need for access controls
- It reduces storage cost for infrequently accessed data Correct answer.
- It guarantees lower retrieval latency
Explanation
Colder storage tiers carry lower per-gigabyte storage prices in exchange for higher retrieval cost or latency, so moving aging data lowers spend. Durability and access control are unaffected by tiering.
Question 29Cloud Computing Fundamentals
Which term describes the ability of a system to continue operating correctly when one of its components fails?
- Fault tolerance Correct answer.
- Elasticity
- Scalability
- Portability
Explanation
Fault tolerance is the property of continuing correct operation despite component failure, usually through redundancy. Scalability and elasticity concern capacity changes, and portability concerns moving workloads.
Question 30Cloud Computing Fundamentals
Which statement best describes the shared responsibility model in public cloud?
- The provider is responsible for every layer of the stack
- Responsibilities are divided equally and identically for every service
- The customer is responsible for every layer of the stack
- Responsibilities are divided between provider and customer depending on the service model Correct answer.
Explanation
Under the shared responsibility model the provider secures the underlying infrastructure while the customer secures what they configure, and the split shifts with the service model. The division is not identical across IaaS, PaaS, and SaaS.
Question 31Cloud Computing Fundamentals
Which metric is most commonly used to measure the availability of a cloud service?
- Number of API calls per second
- Percentage of uptime over a period Correct answer.
- Average object size in storage
- Count of deployed regions
Explanation
Availability is expressed as the proportion of time a service is usable, often quoted as a percentage of uptime. Throughput, object size, and region count do not measure availability.
Question 32Cloud Computing Fundamentals
Which cloud concept refers to a provider's ability to allocate resources from a large shared pool to many customers?
- Multi-tenancy Correct answer.
- Peering
- Orchestration
- Federation
Explanation
Multi-tenancy lets a provider serve many customers from shared pooled infrastructure while isolating their workloads. Federation, peering, and orchestration describe other mechanisms such as identity trust, network linkage, and automation.
Question 33Cloud Computing Fundamentals
Which benefit does a managed message queue provide to a cloud application that must survive spikes in traffic?
- It encrypts all database records
- It guarantees sub-millisecond processing of every message
- It replaces the need for a load balancer
- It decouples producers from consumers so work can be buffered Correct answer.
Explanation
A queue buffers messages so producers can keep publishing during spikes while consumers process at their own rate, decoupling the two tiers. It does not encrypt databases, replace load balancers, or guarantee processing latency.
Question 34Cloud Computing Fundamentals
A company wants to run a stateless web tier that can lose any single instance without user impact. Which design choice best supports this requirement?
- Keep session state in a shared external store and use multiple instances behind a load balancer Correct answer.
- Store session state on each instance's local disk
- Run a single large instance with no redundancy
- Pin each user to one instance for the life of the session
Explanation
Statelessness means any instance can serve any request, so externalizing session state and spreading traffic across instances lets failures be absorbed. Local session state and pinned users tie requests to specific instances.
Question 35Cloud Computing Fundamentals
Which practice best protects a cloud account's root or superuser identity?
- Enabling multi-factor authentication and reserving it for break-glass tasks Correct answer.
- Using it for daily administrative work
- Storing its password in a public repository
- Sharing its credentials with the operations team
Explanation
The root identity has unrestricted access, so it should have multi-factor authentication and be used only when no other identity can perform a task. Routine use or credential sharing multiplies the exposure of the most powerful account.
Question 36Cloud Computing Fundamentals
Which mechanism allows a private subnet's instances to reach the internet for software updates while preventing inbound connections from the internet?
- A virtual private network gateway to the office
- A security group rule allowing all inbound traffic
- A public IP address on each instance
- A NAT gateway Correct answer.
Explanation
A NAT gateway lets instances in a private subnet initiate outbound connections through a public address while blocking unsolicited inbound traffic. Public IPs and permissive inbound rules would expose the instances.
Question 37Cloud Computing Fundamentals
Which approach lets an application store credentials for accessing a database without embedding them in source code?
- A secrets management service that injects credentials at runtime Correct answer.
- Writing them into container image layers
- Hardcoding them in a configuration file committed to the repository
- Emailing them to developers
Explanation
A secrets manager stores credentials centrally, controls access, and can rotate and inject them at runtime so they never appear in code or images. Committing, emailing, or baking secrets into images exposes them broadly.
Question 38Cloud Computing Fundamentals
An application writes logs to local disk on each instance. What is the main risk of this approach in an autoscaling environment?
- Logs are automatically replicated to other regions
- Logs consume too much network bandwidth
- Logs cannot be written by the application
- Logs are lost when instances are terminated Correct answer.
Explanation
Autoscaling replaces instances, and local disk contents disappear with the instance, so logs are lost unless shipped to durable centralized storage. Local disk does not inherently cause network or write failures, nor does it replicate itself.
Question 39Cloud Computing Fundamentals
Which design most reduces the blast radius of a compromised workload credential?
- Granting the workload a narrowly scoped role limited to the resources it needs Correct answer.
- Disabling logging to reduce noise
- Using one shared credential for all workloads
- Granting the workload administrator rights for convenience
Explanation
Least privilege limits what a stolen credential can do, containing damage to the resources that role can reach. Broad rights, shared credentials, and disabled logging all increase the impact and reduce detection.
Question 40Cloud Computing Fundamentals
Which storage characteristic makes object storage a poor fit for a transactional database's data files?
- It cannot store files larger than one megabyte
- It does not support encryption
- It is limited to a single region
- It offers eventual consistency for overwrites and high per-operation latency Correct answer.
Explanation
Object storage is optimized for whole-object reads and writes over HTTP and typically has higher latency and weaker overwrite semantics than block storage, which databases rely on. Object size limits, encryption support, and regional options do not disqualify it in this way.
Question 41Cloud Computing Fundamentals
A team needs a database that remains writable during the loss of one data center. Which configuration best meets this goal?
- A nightly backup stored in the same data center
- A primary node with a synchronous standby in a different data center and automated failover Correct answer.
- A single-node database in one data center
- A read replica in the same rack as the primary
Explanation
A standby in a separate data center with automated failover lets the system promote a new primary when the original site is lost. Single nodes, same-site backups, and same-rack replicas share the failure domain.
Question 42Cloud Computing Fundamentals
Which practice best supports cost accountability across multiple teams sharing one cloud account?
- Consolidating all resources under a single untagged naming convention
- Tagging resources with team and project metadata and reporting on those tags Correct answer.
- Reviewing the bill once a year
- Paying for the largest instance sizes available
Explanation
Cost allocation tags let spend be attributed to the teams or projects that incurred it, enabling chargeback and informed optimization. Untagged resources, infrequent review, and oversized instances obscure or inflate cost.
Question 43Cloud Computing Fundamentals
What is the main purpose of infrastructure as code in a cloud environment?
- To eliminate the need for access control
- To replace the need for any monitoring
- To increase the manual effort of provisioning
- To define infrastructure in versioned files so environments are reproducible and reviewable Correct answer.
Explanation
Infrastructure as code expresses environments as version-controlled definitions that can be reviewed, tested, and reapplied consistently. It complements rather than replaces monitoring and access control.
Question 44Cloud Computing Fundamentals
Which statement best describes the difference between vertical and horizontal scaling in cloud environments?
- Vertical scaling adds servers; horizontal scaling enlarges a server
- Both terms describe moving workloads between regions
- Both terms describe encrypting data at rest
- Vertical scaling enlarges a server; horizontal scaling adds servers Correct answer.
Explanation
Vertical scaling increases the capacity of an existing instance, while horizontal scaling adds more instances to share load. The other choices mislabel the terms or describe unrelated operations.
Question 45Cloud Computing Fundamentals
A team is comparing a serverless function with a continuously running virtual machine for a workload that receives a few requests per hour. Which factor most favors the serverless option?
- Consistent warm execution with no cold starts
- Billing per request and execution time rather than per running hour Correct answer.
- Guaranteed access to the underlying operating system
- The ability to run a persistent background daemon
Explanation
Serverless billing charges for invocations and execution duration, so sparse traffic costs far less than paying for an idle VM around the clock. OS access, daemons, and warm execution are advantages of the VM, not the function.
Question 46Cloud Computing Fundamentals
An application reads the same small reference dataset on every request. Which change most reduces database load without risking stale data beyond a short window?
- Querying the database twice per request
- Caching the dataset in memory with a defined time-to-live Correct answer.
- Storing the dataset only on one instance's local disk
- Removing indexes from the reference table
Explanation
An in-memory cache with a bounded time-to-live absorbs repeated reads and refreshes periodically, trading a short staleness window for much lower database load. Duplicate queries, dropped indexes, and single-instance storage worsen availability or performance.
Question 47Cloud Computing Fundamentals
A workload must process events exactly once even if a consumer crashes after processing but before acknowledging. Which design element is required?
- Disabling message persistence
- Shortening the visibility timeout to zero
- Increasing the consumer count without changing logic
- Idempotent processing keyed on a unique event identifier Correct answer.
Explanation
At-least-once delivery can redeliver a message, so exactly-once effects require the consumer to detect duplicates and process idempotently using a unique key. Scaling consumers, disabling persistence, or zeroing the timeout does not prevent duplicate effects.
Question 48Cloud Computing Fundamentals
Which control most directly limits the damage that a misconfigured public storage bucket can cause?
- Encrypting the bucket with a provider-managed key only
- Enabling versioning on the bucket
- Relying on users to notice the exposure
- Blocking public access at the account level by default Correct answer.
Explanation
An account-level default that blocks public access prevents accidental exposure even when an individual bucket is misconfigured. Detection by users is unreliable, and encryption or versioning does not stop public readability.
Question 49Cloud Computing Fundamentals
A company must prove to auditors that only approved changes reach production infrastructure. Which combination best provides that evidence?
- A single administrator holding all credentials
- Direct console changes made by administrators without logging
- Manual change tickets with no link to deployed resources
- Infrastructure as code with peer review and an audit trail of applied changes Correct answer.
Explanation
Version-controlled definitions reviewed before merge, combined with an audit trail of what was applied, create verifiable evidence that only approved changes reached production. Unlogged console changes, concentrated credentials, and unlinked tickets cannot demonstrate this.
Question 50Cloud Computing Fundamentals
Which reasoning best explains why a multi-region active-active design is more complex than a single-region multi-zone design?
- It must handle data replication conflicts and routing across regions with higher latency Correct answer.
- It eliminates the need to consider data residency
- It removes the need for any failover mechanism
- It guarantees lower cost than a single region
Explanation
Active-active across regions introduces concurrent writes that require conflict resolution and traffic routing over long-distance links with higher latency. It does not remove failover concerns, residency obligations, or cost.