Episode 13 — Cloud Service Models: IaaS, PaaS and SaaS Security Considerations

The purpose of this episode is to compare service models so that you understand how responsibilities and security boundaries shift in the CCSP context. Each model—Infrastructure as a Service, Platform as a Service, and Software as a Service—abstracts different layers of technology. This abstraction defines where the provider’s duties end and the customer’s duties begin. Exam questions often hinge on recognizing these boundaries, since misplacing a responsibility is a frequent cause of real-world incidents. In practice, designing controls without clarity risks either over-controlling areas already managed by providers or, worse, leaving gaps where no one applies protections. By exploring service models in depth, you gain the ability to map controls precisely, ensuring both compliance and resilience.
Service models are standardized delivery abstractions for compute, storage, and application capabilities. At the base, IaaS delivers raw infrastructure resources; PaaS abstracts the infrastructure and delivers managed runtimes; SaaS provides fully managed applications. Each model exists on a spectrum of control, flexibility, and accountability. This standardization allows organizations to choose the balance that suits their needs. For exam purposes, recognizing that these are abstractions helps avoid confusion with marketing labels. In practice, it ensures clear dialogue across providers, auditors, and consumers about what exactly is being purchased and who owns which parts of the security model.
Infrastructure as a Service, or IaaS, delivers virtualized compute, storage, and networking. Customers deploy their own operating systems and applications onto this foundation. The advantage is flexibility: you can recreate nearly any traditional environment in cloud. The trade-off is that more responsibility remains with you. In CCSP preparation, IaaS items often test whether you recognize that infrastructure is managed by the provider but everything on top—the operating system, middleware, and application stack—remains with the customer. This model emphasizes freedom but demands maturity in governance and security operations.
IaaS security responsibilities include hardening and patching operating systems, configuring network controls, and protecting data. Providers ensure physical infrastructure is secure, but customers must manage the guest environments. Exam questions often test whether you understand this distinction: a provider secures the host hypervisor, but the customer must secure the VM. In practice, this means organizations must import traditional IT security disciplines—system patching, configuration management, and monitoring—into the cloud context. IaaS offers flexibility but requires disciplined execution to avoid misconfiguration and exposure.
IaaS network security responsibilities include segmentation, firewall policies, and traffic inspection. Virtual private clouds and subnet designs must be configured by customers. Firewalls must be deployed and tuned to enforce least privilege between segments. Traffic inspection tools, such as intrusion detection or prevention systems, may be integrated to monitor flows. On the exam, IaaS scenarios often hinge on recognizing that these controls are optional and configurable, not enforced by the provider by default. This highlights why many cloud incidents involve misconfigured network boundaries.
Compute security in IaaS relies on baseline images, patching cadence, and drift control. Customers often begin with a “golden image” of a secure operating system, patched and hardened. From there, they must enforce patching schedules to stay current. Configuration drift—the gradual divergence from baseline due to ad-hoc changes—must be controlled with monitoring and automation. Exam items that reference IaaS compute emphasize that customers own this layer, and errors here create vulnerabilities regardless of provider infrastructure quality.
IaaS storage security includes encryption, access policies, and lifecycle management. Providers offer encryption options, but customers decide whether and how to enable them. Access policies define who can read, write, or delete objects and volumes, often integrated with identity and access management systems. Lifecycle management determines how long data is retained, when it is archived, and when it is securely deleted. For CCSP learners, IaaS storage items test whether you recognize that consumer configuration choices drive security outcomes, not the provider’s defaults.
Platform as a Service, or PaaS, shifts abstraction upward. The provider manages operating systems, runtimes, and middleware. Customers focus on code and data. This accelerates development and reduces operational burden but also constrains control. For exam purposes, PaaS scenarios emphasize shared responsibilities: while the provider manages more, customers remain accountable for secure design, data protection, and configuration. The risk is assuming that “managed” means “covered”—missteps here can still cause breaches.
In PaaS, shared responsibilities emphasize secure configuration, identity controls, and data protection. Providers deliver secure runtimes, but customers must define access policies, enforce least privilege, and configure encryption. Exam questions often use PaaS examples to test whether you recognize that identity governance is still the customer’s duty. In practice, many PaaS incidents stem from weak access policies, illustrating that “less responsibility” does not mean “no responsibility.”
Identity and secrets practices in PaaS rely on centralized authentication, well-defined roles, and secure credential handling. Customers should integrate PaaS services into enterprise identity systems, define roles carefully, and use vaults or secret managers instead of embedding keys in code. For CCSP learners, this underscores that even in abstracted environments, identity and credential misuse remains a primary risk.
PaaS data protection combines service-native encryption, secure connections, and controlled service endpoints. Providers often offer encryption by default, but customers remain responsible for managing keys, classifying data, and ensuring that connections use secure protocols like TLS. Restricting endpoints to private networks adds another layer of control. For the exam, PaaS data scenarios emphasize that customer responsibility persists at the data layer even when infrastructure is fully managed.
PaaS runtime governance includes maintaining configuration baselines, applying add-on security controls, and using version pinning to avoid unintended changes. For example, libraries may update automatically unless pinned, creating unexpected shifts in application behavior. Governance ensures stability and security. On the exam, runtime governance items remind you that while the provider operates the platform, the consumer must govern how it is used.
Software as a Service, or SaaS, delivers complete applications managed entirely by the provider. Customers simply configure features, integrate identities, and govern data. This is the most abstracted model, reducing operational burden significantly. However, customer responsibility still exists. On the exam, SaaS scenarios test whether you can identify that while infrastructure and application management are the provider’s duty, customers must secure access and data usage.
SaaS customer responsibilities include integrating with enterprise identity systems, configuring access controls, and enabling security features like multifactor authentication. Customers also govern data classification, retention, and export. Provider assurances cover infrastructure, but tenant configurations determine how secure the application is in practice. Exam questions often highlight SaaS misconfiguration as a consumer accountability, underscoring that customer decisions drive outcomes.
SaaS data protection depends on tenant-defined classification and retention policies. Customers decide what data is stored, for how long, and whether it can be exported or deleted. Providers enforce durability, but customers govern policy. In exam scenarios, SaaS items often hinge on whether you recognize that data governance remains a customer responsibility, even when infrastructure is completely abstracted.
Tenant isolation in SaaS relies on provider architecture, often verified through contractual commitments and assurance reports. Customers must review these documents, such as SOC reports, to gain confidence in isolation mechanisms. While isolation is provider-controlled, customer diligence through contract and audit rights ensures accountability. On the exam, tenant isolation questions emphasize the importance of assurance artifacts and contractual validation.
For more cyber related content and books, please check out cyber author dot me. Also, there are other prepcasts on Cybersecurity and more at Bare Metal Cyber dot com.
Visibility and logging vary across service models, and this has direct impact on monitoring, investigations, and audit readiness. In IaaS, customers have broad visibility into operating systems, networks, and applications, enabling detailed logging but requiring configuration and integration effort. In PaaS, logging is often restricted to platform-provided metrics and service logs, which the customer must consume and interpret. In SaaS, visibility is narrowest, limited to application activity and audit logs that providers choose to expose. For CCSP candidates, the key lesson is that as abstraction increases, so does reliance on provider transparency. On the exam, questions may test whether you recognize that diminished logging in higher-level models increases the need for strong contractual and assurance measures.
Encryption responsibilities also differ. Providers may offer built-in encryption, but customers decide on key ownership models such as Bring Your Own Key (BYOK) or Hold Your Own Key (HYOK). BYOK lets customers import and control keys within the provider platform, while HYOK keeps keys outside the provider’s environment entirely. These models determine custody, compliance posture, and risk appetite. For CCSP learners, exam scenarios often test whether you recognize which party is responsible for key lifecycle activities. In practice, choosing the right model ensures that encryption aligns with legal and governance obligations.
Compliance alignment relies on provider attestations such as ISO certifications or System and Organization Controls (SOC) reports. In IaaS and PaaS, customers must map provider attestations to their own compliance scope, supplementing them with evidence of customer-managed controls. In SaaS, compliance relies heavily on provider documentation and transparency. For the exam, compliance questions emphasize that certifications reduce scope but never eliminate consumer obligations. In professional contexts, aligning compliance ensures both regulatory adherence and stakeholder trust.
Business continuity expectations differ across models. In IaaS, customers design their own RTO and RPO strategies, configuring replication and failover. In PaaS, providers may offer replication by default, but customers must configure backups and validate recovery. In SaaS, continuity is largely managed by the provider, but customers must verify data export, retention, and recovery options. Exam items often test whether you can distinguish where continuity responsibilities sit, highlighting that provider assurances must be matched by consumer planning.
Incident response also requires collaboration. In IaaS, customers gather their own evidence from logs, while providers handle infrastructure-level issues. In PaaS and SaaS, evidence access depends on provider support channels and contractual response timelines. On the exam, incident response questions often emphasize the importance of clear expectations—knowing what evidence you can access, how quickly providers will respond, and how communication flows in a breach scenario.
Penetration testing policies differ as well. In IaaS, customers may perform penetration tests against their own assets, subject to provider approval. In PaaS and SaaS, penetration testing scope is often restricted or prohibited to avoid impacting shared infrastructure. For the CCSP, recognizing these constraints is critical—testing must respect provider rules, and alternatives such as code review or contractual assurances may be required.
Data residency and sovereignty controls are tied to service models. In IaaS, customers select regions for resource deployment. In PaaS, options may be more limited to provider-chosen locations. In SaaS, residency may be fixed unless premium features are purchased. Sovereignty, the legal jurisdiction governing data, follows regardless of residency. Exam scenarios often test whether you understand these differences and their compliance implications.
Multi-cloud and poly-service strategies bring governance complexity. Using multiple providers diversifies risk but introduces inconsistency in controls and monitoring. Poly-service use—combining IaaS, PaaS, and SaaS—further complicates responsibility mapping. For CCSP learners, these strategies highlight the need for governance frameworks that span providers and models.
Cost governance interacts directly with security. Rightsizing resources, purchasing reserved capacity, and choosing feature tiers can influence posture. For instance, enabling advanced encryption or logging may require higher-tier services. Exam questions may test whether you recognize that cost and security are intertwined, and governance must consider both.
Migration strategies—rehost, replatform, and refactor—each have security trade-offs. Rehosting lifts workloads as-is, often carrying legacy risks. Replatforming adapts workloads to cloud-native services, introducing new shared responsibility dynamics. Refactoring rebuilds workloads for cloud-native architectures, maximizing flexibility but requiring deep design changes. On the exam, these patterns often appear in questions about balancing speed with security posture.
Application Programming Interface, or API, limits and throttling affect integration and resilience. Providers enforce rate limits, and mismanagement can cause errors or outages. Security concerns include denial-of-service from excessive requests and proper error handling to prevent data leakage. Exam references to API management highlight the need for secure integration practices in all service models.
Service Level Agreements, or SLAs, codify availability, support, and remedies. In IaaS, SLAs cover infrastructure uptime; in PaaS, platform services; and in SaaS, the application itself. For CCSP learners, SLA questions emphasize aligning risk acceptance with contractual commitments, ensuring business priorities are protected.
Lifecycle management includes awareness of service deprecation, feature changes, and exit strategies. Providers may retire services, forcing migration. Customers must track these changes and plan for portability. For the exam, lifecycle awareness demonstrates maturity in governance and avoids being caught unprepared by provider transitions.
Third-party integration introduces additional trust boundaries. When external services or plug-ins are connected to cloud platforms, customers must perform due diligence, monitor activity, and ensure contracts address security. This extends shared responsibility into a tri-party model: provider, customer, and third party. On the exam, these scenarios emphasize vigilance across all trust boundaries.
In summary, understanding IaaS, PaaS, and SaaS control boundaries is critical for CCSP success. Each model changes the balance of responsibility, visibility, and governance. By mapping these boundaries precisely—recognizing where providers act and where customers remain accountable—you design and operate secure cloud environments. On the exam, this knowledge ensures you identify correct control ownership. In professional practice, it ensures secure, compliant, and resilient operations across all layers of the cloud.

Episode 13 — Cloud Service Models: IaaS, PaaS and SaaS Security Considerations
Broadcast by