CCNA Cloud Concepts, Architecture, and Design Questions

IaaS provides the customer with virtualized compute, storage, and networking while giving them full control over the operating system, middleware, and runtime — exactly what a legacy application needing minimal code changes requires. The customer manages the guest OS and above, while the provider handles the physical infrastructure and hypervisor. This makes IaaS the most appropriate model for lift-and-shift migrations with OS-level control requirements.

Under the shared responsibility model, the customer is responsible for security in the cloud, which includes configuring identity and access management policies and classifying and protecting data. Physical access, hypervisor patching, and physical network fabric are provider responsibilities. Regulated workloads require the customer to focus on data protection and access control while relying on the provider for infrastructure security.

A community cloud is a deployment model where infrastructure is shared by several organizations with common concerns (e.g., compliance, security, jurisdiction). In this scenario, multiple agencies with common compliance requirements share the infrastructure while each retains control over their own resources, which aligns exactly with the community cloud model. This model balances cost-efficiency with the specific needs of a defined community.

Platform as a Service (PaaS) provides the customer with the ability to deploy and run custom applications using the provider's infrastructure, where the customer manages the applications and data, but does not manage the underlying operating system or hardware. This matches the description exactly.

Federated identity requires protocols that carry authentication assertions and user attributes across security domains. SAML 2.0 and OpenID Connect both do this, allowing partner employees to authenticate with their home identity provider and access the shared SaaS application without local accounts. IPsec, DLP endpoint agents, and DHCP snooping operate at network or data layers and cannot federate identities.

An SLA of 99.999% availability (often called 'five nines') allows for a maximum of 5.26 minutes of downtime per year. This is calculated by taking the total minutes in a year (365 days × 24 hours × 60 minutes = 525,600 minutes) and multiplying by the allowed unavailability (100% - 99.999% = 0.001%, or 0.00001 as a decimal). 525,600 × 0.00001 = 5.256 minutes, which rounds to 5.26 minutes.

The public cloud model is designed to provide on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service. These characteristics allow the media streaming company to automatically scale resources during live events and reduce costs when demand is low. Private, community, and hybrid models do not offer the same combination of rapid elasticity and cost efficiency for unpredictable workloads.

The core risk is that data can only be extracted in a proprietary binary format, which prevents migration to another provider. The strongest mitigation is a contractual right to receive the data in a documented, non-proprietary format on demand. Authentication hardening, backups through the same API, and API documentation do not remove the format dependency, so they cannot resolve the portability problem.

Standardizing on Docker images and Kubernetes orchestration with open-source tooling (C) best ensures portability because Docker images are OCI-compliant and Kubernetes is a CNCF-graduated project supported by all major cloud providers. This combination avoids proprietary APIs and allows workloads to be moved between AWS EKS, Azure AKS, Google GKE, or on-premises clusters with minimal changes. Open-source tooling further reduces lock-in by providing consistent APIs and configuration formats.

NIST SP 800-145 defines rapid elasticity as the capability to elastically provision and release resources, in some cases automatically, to scale rapidly outward and inward commensurate with demand. The phrase 'rapidly and elastically provision and release resources, often automatically' maps directly to this characteristic. It is one of the five essential characteristics of cloud computing alongside on-demand self-service, broad network access, resource pooling, and measured service.

In public cloud IaaS, the provider secures facilities, hardware, and the virtualization platform, while the customer secures the guest operating system and its own identity and access management configurations. Guest OS patching and IAM policy management are therefore customer responsibilities, whereas physical facilities, hypervisor maintenance, and physical network redundancy remain with the provider.

Using open standards and APIs ensures that workloads can be moved between providers. Using proprietary APIs, single provider services, or managed services increases lock-in.

Recovery objectives are validated by measurement, not by documentation or provider guarantees. Scheduled failover exercises reveal the real elapsed time to restore service and the true data loss window, accounting for DNS time-to-live, database promotion, and application reconnection. Replication and agreements enable recovery, but only testing produces the evidence that the stated recovery time objective is achievable.

The public cloud model is defined by provider ownership, multi-tenant shared infrastructure, self-service provisioning, and consumption-based billing. Those traits deliver the lowest barrier to entry and allow the startup to stop paying when the product ends. Private and community clouds require dedicated or consortium arrangements, while hybrid assumes a pre-existing private footprint to integrate.

Portability is the design principle concerned with the ability to move workloads, data, and applications between cloud providers or back on-premises without significant re-architecture. It emphasizes avoiding proprietary lock-in through open standards, containers, and abstraction layers. The scenario's phrase 'move workloads between cloud providers or back on-premises without significant re-architecture' is the definition of portability.

Resource pooling means physical infrastructure is shared, so mitigations must either make the data unreadable to anyone outside the tenant or make the logical boundaries between tenants enforceable and auditable. Externally held encryption keys protect confidentiality even if a boundary fails, and virtual network segmentation with distinct identities constrains reachability. Contractual guarantees, disabled logging, and default provider-managed encryption do not change the underlying sharing risk.

SOC 2 Type II is specifically designed to provide assurance over security, availability, processing integrity, confidentiality, and privacy (the Trust Services Criteria). It includes an independent auditor's opinion on the effectiveness of controls over a period. This makes it the most comprehensive for the listed areas.

In a PaaS model, the cloud provider manages the underlying infrastructure, including servers, operating systems, and runtime environments, while the consumer is responsible for the security of their applications and data. This includes configuring application-level security controls and managing user identities and access within the application. These two areas fall under the consumer's control and are critical for protecting the application from threats.

Direct evidence comes from independent testing and from a structured, documented assessment conducted by the enterprise itself. An independent third-party audit report covers defined criteria over a stated period and discloses exceptions, while a framework-mapped questionnaire reviewed with the provider turns vague assurances into specific, verifiable answers. Marketing brochures, terms of service, and reference calls may inform the decision, but they do not establish that the provider's controls are designed and operating effectively.

The HIPAA Security Rule requires covered entities and their business associates to implement administrative, physical, and technical safeguards to ensure the confidentiality, integrity, and availability of ePHI. Before signing a BAA, the healthcare organization must verify that the CSP has these safeguards in place. While data residency, contractual clauses, and SLAs are relevant, they do not replace the need for comprehensive security controls.

The community cloud model is defined by exclusive use among a group of organizations sharing common mission, security, policy, and compliance interests. Because several agencies jointly consume the environment under a shared governance framework while each controls its own data, this matches the community cloud definition precisely rather than single-tenant private, open public, or cross-model hybrid arrangements.

A hybrid cloud is designed to integrate a private cloud with public cloud services so workloads can move or burst between them. Because the company wants to retain its private editing workflows while temporarily using public resources during peaks, the hybrid model provides the required elasticity without forcing a full migration or redesign.

Elasticity is the cloud design principle that allows resources to be automatically and dynamically scaled up or down to match changing demand, typically implemented via auto-scaling groups, horizontal pod autoscalers, or serverless concurrency. It is what lets a workload add capacity during peak load and release it during idle periods, optimizing both performance and cost. Auto-scaling groups are the canonical implementation of elasticity on IaaS platforms.

Using a VPC per tenant provides network-level isolation, which is critical in multi-tenant environments to prevent cross-tenant traffic and unauthorized access. Each VPC acts as a virtual network boundary, with its own IP range, subnets, route tables, and security groups. This logical separation helps enforce security policies and reduces the attack surface, making it the primary security benefit.

Portability is the design principle specifically concerned with avoiding vendor lock-in by ensuring workloads, data, and configurations can be moved between cloud providers with minimal rework. It drives the use of open standards, containerization, and abstraction layers so that an application is not tied to proprietary APIs or services. Reversibility is related but focuses on the ability to exit a provider and bring data back, whereas portability is about the ability to move and run elsewhere.

Private cloud is dedicated to a single organization, providing exclusive use of infrastructure. Public cloud is shared, community is shared by multiple organizations with common interests, and hybrid combines models.

Under the shared responsibility model for IaaS, the customer is responsible for security in the cloud, which includes configuring network controls (security groups, network ACLs) and encrypting data at rest with customer-managed keys. The provider is responsible for security of the cloud, including physical security, hypervisor patching, and physical network redundancy. Therefore, the customer's responsibilities are configuring security groups and network ACLs, and encrypting application data at rest.

IaaS is the only model that gives the customer control over the guest operating system and runtime while the provider manages the physical infrastructure and hypervisor. PaaS, SaaS, and FaaS abstract away the OS layer, removing the control the media company needs. The scenario explicitly requires retaining patching and runtime configuration, which maps directly to IaaS responsibilities.

In the IaaS shared responsibility model, the cloud provider secures the physical infrastructure, hypervisor, and network fabric, while the customer is responsible for everything from the guest OS upward — including OS patching, middleware, runtime, applications, and data. Patch management of the guest operating system (C) is therefore squarely the customer's duty. This is a defining characteristic of IaaS versus PaaS or SaaS, where the provider assumes more of the stack.

The workload is stateless, tolerant of interruption, and cost-sensitive, which maps precisely to spare-capacity compute sold at a discount with short-notice reclamation. Committed-term reservations, physically isolated hosts, and full-price on-demand capacity all fail at least one stated constraint, either cost, availability guarantees, or the eviction behavior the architects are designing around.

The essential characteristics describe on-demand self-service, broad network access, resource pooling, rapid elasticity, and measured service. Rapid elasticity and measured service together capture how capacity flexes with demand and how consumption is metered and reported, which are definitional traits. Residency guarantees, single-tenancy mandates, and perpetual licensing are commercial or deployment choices, not defining cloud characteristics.

An SLA of 99.99% availability allows 0.01% downtime per year. A year has 365 days × 24 hours × 60 minutes = 525,600 minutes. 0.01% of 525,600 is 52.56 minutes, so the maximum allowable downtime is 52.56 minutes per year. This is the correct calculation.

The agency needs a measurable availability commitment plus a defined remedy, and that is precisely what a service level agreement provides. It sets the uptime percentage, the measurement window, exclusions, and the service credits owed when targets are missed. Audit reports, disaster recovery test summaries, and architecture whitepapers are useful evidence about controls and design, but none of them creates an enforceable service commitment with compensation terms.

SaaS (Software as a Service) is the cloud service model where the provider manages the entire stack — infrastructure, operating system, runtime, middleware, and the application itself — leaving the customer responsible only for their data and user access management. Examples include Microsoft 365, Salesforce, and Google Workspace, where the customer configures users and manages data but never touches the underlying platform.

In hybrid cloud, consistent security policies must apply across both environments, and secure connectivity (e.g., VPN or dedicated connection) is essential to protect data in transit and prevent leakage.

A key benefit of hybrid cloud is the ability to keep sensitive data on-premises while leveraging public cloud for less sensitive workloads, thus meeting compliance and security requirements while gaining cloud scalability. This flexibility is a primary driver for hybrid adoption.

Surviving a full region failure requires resources and data in at least two independent regions. Asynchronous database replication avoids the latency and availability penalties of synchronous cross-region writes, cross-region object storage replication protects unstructured data, and DNS failover provides a practical traffic redirection mechanism with acceptable recovery time objectives.

A public cloud is provider-owned and offers on-demand scaling and global availability, which matches the retailer's need to avoid infrastructure ownership while supporting stores in multiple countries and handling seasonal demand spikes. The other models either require dedicated infrastructure, shared governance, or unnecessary private integration.

In SaaS, the provider manages the application, runtime, middleware, operating system, and infrastructure, while the customer consumes the software, often through a browser. Email and productivity suites delivered this way fit the SaaS model precisely, with the customer retaining only limited configuration and user management responsibilities.

Containers bundle the application code together with its libraries, runtime, and configuration into a single immutable image, so the same image runs identically on a developer laptop, on-premises servers, or any cloud VM/container service. This decoupling from the underlying host OS and infrastructure is precisely what delivers portability across cloud environments.

NIST SP 800-145 defines the private cloud deployment model as infrastructure provisioned for exclusive use by a single organization comprising multiple consumers (e.g., business units). The key characteristic is exclusive use by one organization, whether owned/operated by the organization or a third party, and whether on-premises or off-premises.

The workload is interruption-tolerant because it checkpoints frequently and can resume elsewhere, which is exactly the profile that reclaimable spare-capacity pricing is built for. Using infrastructure as a service preserves control over the transcoding runtime while allowing the team to bid on spare capacity and cut cost. Committed discounts, per-seat subscriptions, and dedicated hosts all price or isolate capacity in ways that do not match unpredictable, bursty batch processing.

The requirement combines network isolation, customer-defined addressing and traffic control, and private access to a provider service. A virtual private cloud supplies the isolated network and its subnets, route tables, and gateways; security groups and network ACLs provide the traffic control; and a service endpoint keeps storage traffic on the provider backbone. Content delivery, web application firewalls, and hardware security modules address latency, application attacks, and key protection, not network isolation.

A SOC 2 Type II report attests to the design and operating effectiveness of a service organization's controls over a specified period (typically 3–12 months). It goes beyond a Type I report, which only evaluates control design at a point in time, by testing whether controls operated effectively throughout the audit period.

Kubernetes is an open-source container orchestration platform whose API and manifests (pods, deployments, services) are portable across cloud providers, making it the de facto interoperability standard for containerized workloads. It is governed by the CNCF and implemented by AWS EKS, Azure AKS, GCP GKE, and on-prem distributions, enabling workload portability.

Rapid elasticity is the NIST-defined essential characteristic that allows cloud resources to be provisioned and released automatically, scaling outward and inward commensurate with demand. It enables the cloud consumer to scale capabilities elastically, often in near real-time, without human intervention. This characteristic directly addresses the ability to scale up and down rapidly based on demand, as stated in the question.

On-demand self-service is the essential cloud characteristic that enables a consumer to unilaterally provision computing capabilities, such as server time and network storage, automatically without requiring human interaction with each service provider. This is exactly what the question describes: automatic provisioning without human interaction. Rapid elasticity, broad network access, and resource pooling are related but distinct characteristics that do not specifically address the automatic provisioning aspect.

On-demand self-service is the NIST-defined characteristic where a consumer can unilaterally provision computing capabilities, such as server time and storage, automatically without human interaction with the provider. This matches the question's wording exactly.

99.95% uptime over 365 days allows 0.05% downtime. 365 days is 525,600 minutes; 0.05% of 525,600 is 262.8 minutes. Therefore the maximum allowed downtime is 262.8 minutes.

PaaS provides a managed platform where the provider handles the underlying infrastructure, operating system, and middleware (runtime, databases, web servers), while the customer manages only the application and data. This exactly matches the requirement that the company manages only the application and data. Examples include AWS Elastic Beanstalk, Google App Engine, and Azure App Service.

Portability is the cloud design principle specifically concerned with avoiding provider lock-in by designing workloads so they can be moved between cloud providers with minimal re-engineering. It emphasizes use of open standards, containerization, abstraction layers, and portable data formats so that the same workload can run on AWS, Azure, or GCP without significant refactoring. This directly matches the scenario of migrating workloads across providers without significant re-engineering.

The scenario requires both exclusive control over the underlying infrastructure and the ability to extend into elastic public capacity on demand. A hybrid cloud is the only model that pairs a controlled private environment with public cloud bursting, satisfying the latency-sensitive rendering requirements while avoiding permanent over-provisioning of owned hardware.

A community cloud is shared infrastructure provisioned for exclusive use by a specific community of consumers from organizations that have shared concerns (e.g., mission, security requirements, policy, compliance). Several government agencies with similar security and compliance requirements is the canonical example, because they can share costs while meeting common regulatory mandates like FedRAMP or CJIS. The defining trait is a bounded community with shared interests, not the general public.

Rapid elasticity is the cloud characteristic that allows resources to scale automatically and dynamically based on demand, including scaling out and in quickly. The requirement for automatic scaling of stateless compute based on real-time demand directly maps to this principle. It ensures capacity matches workload without manual intervention.

IaaS gives the customer control over the operating system, middleware, runtime, and applications while the provider manages the physical host, hypervisor, and networking fabric. This is the highest level of customer control among the four models listed. PaaS, SaaS, and FaaS progressively abstract away more of the stack, reducing customer control.

Consistent security policy across environments is critical in hybrid cloud because workloads and data span on-premises and cloud boundaries, and inconsistent controls create gaps attackers can exploit. Unified policy ensures identity, encryption, logging, and access controls are applied uniformly. This is a foundational governance requirement for hybrid architectures.

Assessing data residency and audit obligations requires verifiable evidence about where data lives and how controls are validated. A data flow diagram identifies storage, processing, and replication locations, while independent audit reports and certifications demonstrate that controls have been examined against recognized standards. Together they give the architect the factual basis needed for compliance due diligence.

Portability is the design principle of avoiding dependence on a specific vendor's proprietary services so workloads can move between environments. Using containers orchestrated by Kubernetes is a classic portability strategy because container images and Kubernetes manifests are largely vendor-neutral. This directly addresses the goal of minimizing vendor lock-in.

PaaS (Platform as a Service) provides a managed runtime environment where consumers deploy applications built with provider-supported languages, libraries, and tools, while the provider manages the underlying servers, storage, networking, and OS. The consumer controls only the deployed application and its configuration, not the infrastructure — exactly matching the scenario described.

FaaS is the most serverless option, allowing developers to run code in response to events without provisioning or managing servers. It aligns with the startup's goal of minimizing infrastructure management and focusing on code development. The provider handles all scaling, patching, and capacity planning, making it ideal for teams with limited operational resources.

In SaaS the provider secures everything from the physical facility up through the application, while the consumer retains control over its own identities, access entitlements, and the data it chooses to place in the service. Physical security, hypervisor maintenance, and guest operating system patching all sit beneath the application and never move to the subscriber.

The 99.99% SLA allows for a maximum of 0.01% downtime per month. Assuming a 30-day month (43,200 minutes), 0.01% of 43,200 minutes is 4.32 minutes. Therefore, the correct answer is 4.32 minutes.

SaaS is the model where the provider manages the entire application stack, and the customer only handles configuration and user management. The scenario explicitly states the provider manages the application, servers, and operating system, leaving only account and settings configuration to the business. IaaS, PaaS, and FaaS all require more customer responsibility for the application or runtime.

PaaS is designed to abstract away servers, operating systems, and runtime management while still allowing custom application deployment. The startup wants to write code and let the provider handle patching and scaling, which is exactly the division of labor PaaS provides, unlike IaaS where infrastructure remains the customer's concern or SaaS where no custom code is deployed.

On-demand self-service is the cloud characteristic defined by NIST SP 800-145 that allows a consumer to unilaterally provision computing capabilities, such as server time and network storage, automatically without requiring human interaction with each service provider. This exactly matches the scenario of automatic provisioning without provider interaction.