Sign in
Home/EC-Council/Certified Cloud Security Engineer (CCSE)/Free questions

Certified Cloud Security Engineer (CCSE) — Free Practice Questions

10 free sample questions from a bank of 40, with the correct answers and explanations. No signup required — start practising right now.

1Cindy Williams has been working as a cloud security engineer in an IT company situated in Austin, Texas. Owing to the robust security and cost-effective features provided by AWS, her organization adopted AWS cloud-based services. Cindy has deployed an application in the Amazon Elastic Compute Cloud (EC2) instance. Which of the following cloud computing service model does the Amazon EC2 instance represent?
  • SaaS
  • DaaS
  • PaaS
  • IaaS
Answer: D

Answer: D — IaaS (Infrastructure as a Service)

The short version

Amazon EC2 gives you virtual machines — the compute, plus the block storage and networking attached to them. You manage the guest OS, the patch level and the applications; AWS manages the physical hosts, power, cooling and the physical network. That is the textbook IaaS boundary.

Key concepts

  • Shared responsibility in cloud — the provider secures the infrastructure beneath the abstraction; the customer secures everything they deploy on top of it.
  • Compute responsibility split — in IaaS the customer owns the guest operating system, its patching, identity controls and network configuration inside the instance.
  • Shared-tenure risk — the physical host is multi-tenant, so side-channel and hypervisor-escape risks sit with the provider, while instance-to-instance isolation is enforced by the provider's virtualisation layer.
  • Cloud storage/database security — EBS volumes, snapshots and instance profiles with scoped IAM roles carry the same encryption-at-rest and least-privilege obligations as any other cloud asset.

Why D is correct

The three service models differ in how much of the stack the customer owns. With EC2 the customer chooses the instance image, the OS version, the security groups and the processes running inside — control that only exists in IaaS. AWS retains everything below the virtualisation layer: the metal, the hypervisor, the host network and the physical data centre. That "customer owns OS up, provider owns OS down" split is exactly the definition of Infrastructure as a Service, so D is correct.

Why the others are wrong

  • A. SaaS — the customer consumes a finished application over the network and owns none of the stack. There is no OS to patch and no instance to secure, because the provider runs everything. Deploying an application on EC2 is the opposite of SaaS.
  • B. DaaS — Desktop as a Service delivers a managed desktop session in a data centre; the provider owns the OS and the patching. It is a specific SaaS-like delivery model, and it is still not what an EC2 instance is.
  • C. PaaS — the provider would own the OS, the runtime and the patch level, exposing only the application and its data. Choosing an EC2 instance means deliberately taking on the OS and runtime responsibilities that PaaS would have absorbed.

CCSE exam tip

Shared-responsibility questions are fastest when you locate the cut line: in IaaS the customer owns the guest OS and everything above it; in PaaS the customer owns the application and data; in SaaS the customer owns only data and configuration.

2The TCK Bank adopts cloud for storing the private data of its customers. The bank usually explains its information sharing practices to its customers and safeguards sensitive data. However, there exist some security loopholes in its information sharing practices. Therefore, hackers could steal the critical data of the bank’s customers. In this situation, under which cloud compliance framework will the bank be penalized?
  • ITAR
  • GLBA
  • NIST
  • GDPR
Answer: B

Answer: B — GLBA (Gramm-Leach-Bliley Act)

The short version

GLBA is the US financial-services law that carries enforceable Safeguards Rule obligations for institutions holding customer non-public personal information. TCK Bank's failure to safeguard customer data and its weak disclosure of sharing practices is a GLBA problem.

Key concepts

  • GLBA Safeguards Rule — requires financial institutions to develop, implement and maintain an information-security programme with a written risk assessment, access controls, encryption and monitoring.
  • Privacy vs safeguarding — the Privacy Rule governs what may be disclosed and requires an opt-in notice for non-affiliated third parties; the Safeguards Rule governs how data is protected while held.
  • Third-party sharing risk — onward disclosure to affiliates and service providers is permitted by default, which is exactly where information-sharing loopholes create exposure.
  • Penalties and attestation — GLBA enforcement includes examinations, and the FTC Safeguards Rule requires a qualified individual to attest annually to the board.

Why B is correct

The scenario is a US bank holding private customer data. GLBA is the statute that both mandates safeguards for customer information held by financial institutions and regulates how that information may be shared with non-affiliated third parties, with the FTC able to penalise non-compliance under the Safeguards Rule and Privacy Rule. The stem's two failure modes — inadequate protection of sensitive data and deficient disclosure of sharing practices — map directly onto those two halves of the statute.

Why the others are wrong

  • A. ITAR — the International Traffic in Arms Regulations govern the export and re-export of defence articles, technical data and defence services. A bank's customer records are not controlled technical data under DDTC jurisdiction, so ITAR penalties are inapplicable.
  • C. NIST — NIST SP 800-series publications are voluntary control catalogues and the basis of the Safeguards Rule, but NIST itself does not conduct enforcement and cannot fine a bank. Naming the framework when the question asks under which compliance framework the bank will be penalised picks the wrong instrument.
  • D. GDPR — the EU regulation with real penalties, but TCK Bank is described in generic terms with no EU establishment or EU data subjects. GLBA is the applicable US law for a US financial institution's non-public personal information.

CCSE exam tip

Separate three concepts that get confused: a law that penalises you (GLBA, HIPAA, GDPR, ITAR), a framework you map controls to (NIST, CIS), and a certification (ISO 27001). A question using the words "will be penalised" wants the law.

3Tom Holland works as a cloud security engineer in an IT company located in Lansing, Michigan. His organization has adopted cloud-based services wherein user access, application, and data security are the responsibilities of the organization, and the OS, hypervisor, physical, infrastructure, and network security are the responsibilities of the cloud service provider. Based on the aforementioned cloud security shared responsibilities, which of the following cloud computing service models is enforced in Tom’s organization?
  • Software-as-a-Service
  • On-Premises
  • Infrastructure-as-a-Service
  • Platform-as-a-Service
Answer: C

Answer: C — Infrastructure-as-a-Service (IaaS)

The short version

The split described — customer owns user access, application and data; provider owns OS, hypervisor, physical, infrastructure and network — is the canonical IaaS division of labour. Nothing is left for the customer to build but the layer above the virtualisation boundary.

Key concepts

  • Shared responsibility matrix — the customer's duties shrink as the abstraction rises: IaaS starts at the guest OS, PaaS starts at the application, SaaS starts at data/configuration.
  • Guest OS responsibility — kernel patching, host-based firewalling, file permissions, secure-shell hygiene, logging agents and EDR installation all become customer duties.
  • Provider-side responsibility — hypervisor patching and escape-surface hardening, physical host security, facility access, power and cooling, and backbone networking.
  • Workload protection — with OS-level control retained, the customer must deploy in-guest protection: patching cadence, EDR/host IDS, file-integrity monitoring and CIS-hardened images.

Why C is correct

Enumerate the layers the provider keeps: operating system, hypervisor, physical servers, underlying infrastructure and network. In IaaS the hypervisor boundary means the customer provisions and administers virtual machines, which is precisely the stack from the guest OS upward — identity and access, the application and its data. That matches the scenario exactly, so the enforced model is IaaS.

Why the others are wrong

  • A. Software-as-a-Service — SaaS hands the customer a finished application; the provider owns the OS, hypervisor and application platform. The scenario explicitly places the OS with the provider but leaves user access, application and data with the customer, which SaaS would not.
  • B. On-Premises — in an on-premises estate the customer owns everything, including the hypervisor and physical layer. Here the customer explicitly does not own the OS, hypervisor, physical or network layers, so this is not on-premises.
  • D. Platform-as-a-Service — PaaS would leave the guest OS, runtime and patch management with the provider, and the customer would deploy only application code and data. The scenario's inclusion of user access management and OS ownership with the customer puts it below PaaS.

CCSE exam tip

Anchor on the single most diagnostic line in the shared-responsibility matrix: if the customer owns the guest OS, the model is IaaS. If the customer owns only the app and data, it is PaaS; only data and configuration, SaaS.

4An organization, PARADIGM PlayStation, moved its infrastructure to a cloud as a security practice. It established an incident response team to monitor the hosted websites for security issues. While examining network access logs using SIEM, the incident response team came across some incidents that suggested that one of their websites was targeted by attackers and they successfully performed an SQL injection attack. Subsequently, the incident response team made the website and database server offline. In which of the following steps of the incident response lifecycle, the incident team determined to make that decision?
  • Containment
  • Analysis
  • Coordination and information sharing
  • Post-mortem
Answer: A

Answer: A — Containment

The short version

The team took the affected website and database server offline. Taking a system out of service to stop the bleeding is containment — limiting the blast radius after the incident is confirmed and analysis has identified the affected assets.

Key concepts

  • Incident response lifecycle — preparation; identification/analysis; containment; eradication and recovery; post-mortem/lessons learned.
  • Containment vs eradication — containment limits ongoing damage (isolate, take offline, block, revoke); eradication removes the root cause (patch the SQLi, remove the implant).
  • Scope isolation — network segmentation, host isolation and access revocation keep a compromised asset from becoming a pivot into the rest of the estate.
  • Evidence preservation — taking a system offline must be done so logs, memory artefacts and the compromised database state remain available for forensic analysis and scoping the data breach.

Why A is correct

The decision described is to make the affected website and database server offline. That is a classic containment action: it stops the attacker maintaining access and halts further data exfiltration, without yet fixing the vulnerability or restoring service. Containment sits immediately after analysis in the response lifecycle, and its explicit objective is to limit damage while eradication and recovery are planned — which matches "made the website and database server offline" precisely.

Why the others are wrong

  • B. Analysis — analysis is the investigative phase: confirming the SQL injection, determining what was accessed or exfiltrated and scoping affected records from the SIEM and web logs. The team had already finished that work; the stem says they then decided to take the assets offline.
  • C. Coordination and information sharing — this is the notification and stakeholder phase: legal, privacy, regulators, customers, law enforcement, threat intelligence. No such activity is described.
  • D. Post-mortem — the post-mortem is the retrospective after recovery, in which the team documents root cause, timeline and lessons learned. It produces documentation and process changes, not an outage.

CCSE exam tip

Sort IR actions by verb: isolate or disable = containment, patch and remove = eradication, restore traffic = recovery, document and review = post-mortem. "Took it offline" is the containment verb.

5Global CloudEnv is a cloud service provider that provides various cloud-based services to cloud consumers. The cloud service provider adheres to the framework that can be used as a tool to systematically assess cloud implementation by providing guidance on the security controls that should be implemented by specific actors within the cloud supply chain. It is used as the standard to assess the security posture of organizations on the Security, Trust, Assurance, and Risk (STAR) registry. Based on the given information, which of the following cybersecurity control frameworks does Global CloudEnv adhere to?
  • CDMI
  • CSA CAIQ
  • CSA CCM
  • ITU-T X.1601
Answer: C

Answer: C — CSA CCM (Cloud Security Alliance Cloud Controls Matrix)

The short version

The description matches the Cloud Controls Matrix exactly: a framework that maps security controls to specific actors across the cloud supply chain, and the assessment instrument used for the Cloud Security Alliance's STAR Registry.

Key concepts

  • CSA CCM — a comprehensive control catalogue organised by control domains and mapped to the shared-responsibility actors: cloud consumer, cloud service provider, cloud carrier, and the physical/virtualisation/operations layers in between.
  • STAR Registry — the CSA's public registry of cloud providers' completed CCM self-assessments (STAR-Lite and full STAR), which customers use for due diligence and vendor risk assessment.
  • Control mapping — CCM domains include IAM, data security and privacy, infrastructure, operations, incident response, and supply-chain management.
  • Vendor risk and third-party assurance — STAR-level publication turns a self-assessment into comparable evidence for procurement and third-party risk programmes.

Why C is correct

Three identifying details converge on CCM: it systematically assesses cloud implementation, it assigns controls to specific actors within the cloud supply chain (consumer, provider, carrier), and it is the standard used to assess security posture for the STAR registry. The CCM's whole design goal is to make control ownership explicit across those actors so that gaps at each handoff are visible. That is a description of CCM and nothing else.

Why the others are wrong

  • A. CDMI — the Cloud Data Management Interface is an interface standard for managing data stored in cloud storage services (naming, capacity, discovery), not a security-control assessment framework.
  • B. CSA CAIQ — the Consensus Assessment Initiative Questionnaire is the evidence questionnaire that accompanies a CCM assessment; it is what a provider answers, not the catalogue of controls itself. It is the nearest neighbour and therefore the most attractive distractor.
  • D. ITU-T X.1601 — a taxonomy of cloud computing terms and definitions. Useful vocabulary, but a taxonomy assigns no controls to actors and produces no assessment posture, so it cannot be the framework described.

CCSE exam tip

For cloud assurance questions, map the description: "controls mapped to supply-chain actors + STAR registry" means CSA CCM; "questionnaire a provider fills in" means CAIQ; "definitions of cloud terms" means ITU-T X.1601.

6Global SoftTechSol is a multinational company that provides customized software solutions and services to various clients located in different countries. It uses a public cloud to host its applications and services. Global SoftTechSol uses Cloud Debugger to inspect the current state of a running application in real-time, find bugs, and understand the behavior of the code in production. Identify the service provider that provides the Cloud Debugger feature to Global SoftTechSol?
  • Google
  • IBM
  • Azure
  • AWS
Answer: A

Answer: A — Google

The short version

Cloud Debugger is a Google Cloud Platform capability for inspecting a running application in production: capture method-level execution state, profiles and variable values without restarting or redeploying. It is a GCP service.

Key concepts

  • Cloud Debugger — zero-configuration production debugging that reads live process state via the debug agent, with configurable breakpoints and dynamic variable/logic inspection.
  • Production-safe inspection — attaching a debugger to a running production instance without a redeploy, which is the point of the feature.
  • Cloud APM / monitoring — the same class of capability exists natively elsewhere (Azure Application Insights and Snapshot Debugger, AWS X-Ray), which is why the provider question is legitimate rather than a giveaway.
  • Workload protection and DevSecOps — inspecting live application behaviour in production is part of securing workloads and of shift-left/continuous assurance practice.

Why C is correct

The described capability — inspect the current state of a running application in real time, find bugs, understand behaviour in production — is Cloud Debugger, and Cloud Debugger is a Google Cloud service, part of Google Cloud's operations and debugging tooling. The stem names the feature explicitly, so identifying the provider follows directly from the feature.

Why the others are wrong

  • B. IBM — IBM Cloud (with Red Hat OpenShift) offers its own diagnostics tooling, but not a service named Cloud Debugger with this behaviour. Name-matching is decisive: the product named in the stem is Google's.
  • C. Azure — Microsoft Azure's equivalent is Application Insights, which includes the Snapshot Debugger feature for production diagnostics. Close in purpose, wrong product name.
  • D. AWS — AWS offers X-Ray for distributed tracing and service maps, plus CloudWatch for logs and metrics. X-Ray is distributed-trace analysis, not interactive inspection of a live application's method-level state.

CCSE exam tip

Learn the debugging-tool map: GCP Cloud Debugger, Azure Application Insights / Snapshot Debugger, AWS X-Ray. These three are repeatedly used as "which provider" answers in cloud exam sets.

7Samuel Jackson has been working as a cloud security engineer for the past 12 years in VolkSec Pvt. Ltd., whose applications are hosted in a private cloud. Owing to the increased number of users for its services, the organizations is finding it difficult to manage the on-premises data center. To overcome scalability and data storage issues, Samuel advised the management of his organization to migrate to a public cloud and shift the applications and data. Once the suggestion to migrate to public cloud was accepted by the management, Samuel was asked to select a cloud service provider. After extensive research on the available public cloud service providers, Samuel made his recommendation. Within a short period, Samuel along with his team successfully transferred all applications and data to the public cloud. Samuel’s team would like to configure and maintain the platform, infrastructure, and applications in the new cloud computing environment. Which component of a cloud platform and infrastructure provides tools and interfaces to Samuel’s team for configuring and maintaining the platform, infrastructure, and application?
  • Virtualization Component
  • Computer Component
  • Physical and Environment Component
  • Management Component
Answer: D

Answer: D — Management Component

The short version

Within the Cloud Security Alliance CCM, the Management component is the domain covering governance and the people, processes and procedures that configure, operate and maintain the cloud platform and its services — exactly what a team does after migrating off its own data centre.

Key concepts

  • CSA CCM components — Management, People, Processes, Physical, Virtualization, Workload, Operations, Data, Application, Identity & Access, Network.
  • Management component — governance and control activities: organisational roles and responsibilities, service management, asset and configuration management, change management, and ongoing management of the platform.
  • People vs Processes vs Management — People covers organisational and HR controls (roles, training, background checks); Processes covers the operational SDLC and change procedures; Management covers the authority and mechanisms used to direct and maintain the service.
  • Migrating to public cloud — moving from an owned data centre to a public cloud transfers the physical, virtualisation and much of the operations burden to the provider, leaving governance and configuration management as the customer's core duty.

Why D is correct

The scenario describes a team whose remit is to configure and maintain the platform, the infrastructure and the applications after the migration. Those are management activities in CCM terms: defining and running the configuration of cloud services, managing changes and administering the estate. The physical and virtualisation layers, by contrast, are now the provider's responsibility once the workload leaves VolkSec's own data centre — which is precisely why the question is about the Management component rather than the Physical one.

Why the others are wrong

  • A. Virtualization Component — covers hypervisors, host patching, VM escape and tenant isolation. Once workloads run on a public cloud, that layer belongs to the cloud service provider, not to Samuel's team.
  • B. Computer Component — not a CCM component at all; it reads like a generic hardware label rather than a defined domain, and it describes infrastructure the customer no longer operates.
  • C. Physical and Environment Component — covers data-centre facilities, physical access, environmental controls and media handling. All of that is explicitly the provider's responsibility after a public-cloud migration, so it cannot be the team's domain.

CCSE exam tip

For CCM component questions, first ask who owns the layer now. After a public-cloud migration, the customer's residual duty is governance and configuration management — the Management component — while physical, virtualisation and operations move to the provider.

8Global InfoSec Solution Pvt. Ltd. is an IT company that develops mobile-based software and applications. For smooth, secure, and cost-effective facilitation of business, the organization uses public cloud services. Now, Global InfoSec Solution Pvt. Ltd. is encountering a vendor lock-in issue. What is vendor lock-in in cloud computing?
  • It is a situation in which a cloud consumer cannot switch to another cloud service provider without substantial switching costs
  • It is a situation in which a cloud service provider cannot switch to another cloud service broker without substantial switching costs
  • It is a situation in which a cloud consumer cannot switch to a cloud carrier without substantial switching costs
  • It is a situation in which a cloud consumer cannot switch to another cloud service broker without substantial switching costs
Answer: A

Answer: A — A situation in which a cloud consumer cannot switch to another cloud service provider without substantial switching costs

The short version

Vendor lock-in is the consumer's inability to move to a competing provider without prohibitive cost or effort. The asymmetry matters: it is the customer who is locked in, because of data formats, APIs, identity models, contractual terms and the labour of migration.

Key concepts

  • Lock-in drivers — proprietary APIs and SDKs, egress fees, non-portable identity/IAM models, closed data formats, managed-service dependencies and long contracts with exit penalties.
  • Egress and data portability — the ability to export data in an open format at reasonable cost is the main practical mitigation; egress charges make leaving expensive even when the data is yours.
  • Multi-cloud and abstraction — architectural mitigation via containers, open standards, IaC and a neutral identity layer reduce the switching cost.
  • Cloud broker/carrier vs provider — the cloud service provider is the party that supplies the service; the cloud carrier delivers it to the consumer and the broker assists with selection and consumption. Lock-in is experienced by the consumer against the provider.

Why A is correct

The definition must name the party that is constrained and the reason. It is the cloud consumer who cannot move to another cloud service provider because doing so would cost too much in effort, money and time. Options B, C and D each replace either the constrained party (the provider in B) or the counterparty (a broker in D, a carrier in C) with a party that is not the service supplier, which destroys the definition.

Why the others are wrong

  • B. A situation in which a cloud service provider cannot switch to another cloud service broker — reverses the roles. Providers do not depend on brokers for their service capability in the way consumers depend on providers; switching costs are borne by the customer.
  • C. A situation in which a cloud consumer cannot switch to a cloud carrier — a cloud carrier delivers a finished service to the end user and is typically chosen by the consumer or broker, so it is not the party that creates platform lock-in. Lock-in attaches to the underlying provider's technology.
  • D. A situation in which a cloud consumer cannot switch to another cloud service broker — the counterparty is wrong again. A broker is an intermediary that helps pick and consume services; the technical and contractual lock-in lives with the provider.

CCSE exam tip

Define any cloud term by asking three questions: who is constrained, what are they constrained from, and what causes the cost. Lock-in = the consumer, the provider, and migration cost.

9Richard Roxburgh works as a cloud security engineer in an IT company. His organization was dissatisfied with the services of its previous cloud service provider. Therefore, in January 2020, his organization adopted AWS cloud-based services and shifted all workloads and data in the AWS cloud. Richard wants to provide complete security to the hosted applications before deployment and while running in the AWS ecosystem. Which of the following automated security assessment services provided by AWS can be used by Richard to improve application security and check the application for any type of vulnerability or deviation from the best practices automatically?
  • AWS CloudFormation
  • Amazon Inspector
  • AWS Control Tower
  • Amazon CloudFront
Answer: B

Answer: B — Amazon Inspector

The short version

Amazon Inspector is AWS's managed vulnerability-assessment service. It continuously probes your compute, container and Lambda workloads, evaluates them against a database of vulnerabilities and best-practice checks, and reports findings so they can be remediated — automatically and without hosting scanners.

Key concepts

  • Managed vulnerability scanning — the provider runs the scanners and the vulnerability intelligence, so there is no agent fleet or scan appliance for the customer to operate.
  • CVE and CIS best-practice checks — Inspector evaluates instances, container images and Lambda functions against known CVEs and against hardened-baseline configurations.
  • DevSecOps / shift-left — automated assessment in the pipeline and on running infrastructure turns vulnerability management into a continuous control rather than a periodic scan.
  • Remediation and reporting — findings are surfaced through AWS Security Hub and Amazon EventBridge, and can drive remediation via Systems Manager, closing the loop from detection to patch.

Why B is correct

The requirement is an automated security assessment of hosted applications, checking for vulnerabilities and deviations from best practices before deployment and while running, without the customer managing scanners. Amazon Inspector is exactly that service in AWS: continuous, managed assessment of compute resources against a vulnerability database plus security best-practice checks, with findings integrated into Security Hub for prioritised remediation.

Why the others are wrong

  • A. AWS CloudFormation — an infrastructure-as-code service for declarative provisioning of AWS resources from templates. It is central to a DevSecOps pipeline, but it does not scan workloads for vulnerabilities or assess them against best practices.
  • C. AWS Control Tower — a governance and account-management service that provisions a multi-account landing zone, applies guardrails and enables central logging and security services. It enforces controls across accounts; it does not perform per-workload vulnerability assessment.
  • D. Amazon CloudFront — a CDN and edge-accelerated content delivery network, fronted by AWS WAF and Shield for edge security. It protects delivery of content, not the vulnerability posture of the origin application.

CCSE exam tip

Keep the AWS service map straight by function: Inspector = vulnerability assessment, Security Hub = findings aggregation, GuardDuty = threat detection, Config = compliance evaluation, Macie = data classification. A "check my workloads for vulnerabilities automatically" question is Inspector.

10GlobalCloud is a cloud service provider that offers various cloud-based secure and cost-effective services to cloud consumers. The customer base of this organization increased within a short period; thus, external auditing was performed on GlobalCloud. The auditor used spreadsheets, databases, and data analyzing software to analyze a large volume of data. Based on the given information, which cloud-based audit method was used by the auditor to collect the objective evidence?
  • CAAT
  • Re-Performance
  • Striping
  • Gap Analysis
Answer: A

Answer: A — CAAT (Computer-Assisted Audit Technique)

The short version

Auditors who extract large populations of data into spreadsheets, databases and analysis tools and test them programmatically are using a computer-assisted audit technique. CAAT is the standard term for exactly this method of collecting and examining objective evidence.

Key concepts

  • CAAT — any use of computer software by the auditor to read, extract, manipulate or test data as part of the audit; includes database queries, spreadsheet analysis and scripted extraction of full populations.
  • Population testing vs sampling — CAAT lets an auditor test 100% of transactions rather than a small sample, which is both faster and more persuasive.
  • Objective vs subjective evidence — CAAT output is objective, machine-generated evidence; corroboration and professional judgement supply the interpretive element.
  • Other audit methods — re-performance re-executes the client's procedure to test its operation; gap analysis compares current state to a target framework; striping is a data-destruction/erasure technique that fragments writes across multiple passes.

Why A is correct

The scenario describes an external auditor taking a high volume of data into spreadsheets, databases and analysis software to test it. That is the definition of a computer-assisted audit technique: the auditor uses general-purpose software to extract, reconcile and analyse complete data sets rather than manually inspecting a sample. The method is explicitly about collecting and testing objective evidence, which is what CAAT output constitutes, so A is correct.

Why the others are wrong

  • B. Re-Performance — the auditor independently re-executes the client's control (for example, re-running an approval workflow) to test whether it operated consistently. It is a substantive testing technique, not a data-extraction and analysis method, and it does not involve spreadsheets or databases of extracted records.
  • C. Striping — a technique for securely erasing data by overwriting it across several passes so no single pass can be reconstructed. It destroys data rather than collecting evidence, so it has nothing to do with an external audit of volume and accuracy.
  • D. Gap Analysis — comparing the current control environment against a target framework or required state and listing the differences. The scenario describes examining actual transaction data with analysis tools, not comparing a control set to a standard, so gap analysis is the wrong method.

CCSE exam tip

Cloud-audit method keywords: tools extracting and testing data = CAAT; re-executing a control = re-performance; current state vs required state = gap analysis; multi-pass overwrite = striping.

Want the full bank of 40 questions for Certified Cloud Security Engineer (CCSE)? See all practice exams.