6 Cloud Native Test

**1. Which of the following is a recommended practice for defining interfaces in API development? **

1 Defining interfaces after development is complete.
2 Defining interfaces only for current use cases
3 Defining interfaces to support current, planned, and foreseeable uses
4 Defining interfaces without considering reuse.

The correct answer is:

3 — Defining interfaces to support current, planned, and foreseeable uses ✅

In API development, you want your interface design to be forward-looking. That means:

Cover current requirements
Anticipate planned features
Allow for reasonable future expansion without breaking backward compatibility

This approach reduces costly redesigns and avoids breaking changes for API consumers later.

Defining interfaces only for the current state (option 2) or after development is done (option 1) is risky, and skipping reuse considerations (option 4) is a common cause of technical debt.

2 Which of the following is a recommended practice for registering interfaces in API development

1 Registering only interfaces that are currently exposed as endpoints
2 Not registering any interfaces that are not currently exposed as endpoints - 3 Registering all interfaces that are currently exposed as endpoints
4 Registering all interfaces even if not currently exposed as endpoints

The correct answer is:

3 — Registering all interfaces that are currently exposed as endpoints ✅

In API development, registration is about making exposed endpoints discoverable and documented—usually in an API registry, developer portal, or service catalog.

Why not 1 or 2? Those could mean leaving some exposed endpoints unregistered, which is risky because undocumented APIs are harder to maintain and more prone to misuse.
Why not 4? You shouldn’t register endpoints that are not actually exposed—this could cause confusion or security risks.

So the best practice is: 🔹 Register every API interface you expose so it’s documented, searchable, and properly governed, but don’t register internal-only or unused interfaces.

4 Which of the following is a recommended practice for using common context and library services in API development?

1 Ignoring common context and library services
2 Using only some common context and library services
3 Using all common context and library services
4 Using common context and library services only for security

The correct answer is:

3 — Using all common context and library services ✅

In API development, common context and library services (e.g., logging, security, error handling, metrics, authentication, configuration management) are designed to ensure consistency, reusability, and maintainability across APIs.

Using all of them ensures every API benefits from the shared standards and avoids duplication of effort.
Option 1 ignores them entirely — leading to inconsistent implementations.
Option 2 cherry-picks, which can cause gaps in functionality or governance.
Option 4 limits them to security, missing other key benefits like observability, performance, and maintainability.

5 Are applications built according to a data-centric design?

1 Applications are built according to an application-centric design, data is data-centric
2 Applications are built according to a user-centric design, data is data-centric
3 Applications are built according to a data-centric design
4 It depends on the type of application

The correct answer is:

1 — Applications are built according to an application-centric design, data is data-centric ✅

In architecture principles:

Applications are generally designed around specific business functions and use cases → application-centric.
Data is designed, modeled, and governed in a data-centric way → independent of any single application so it can be reused and shared.

This distinction is important because while applications may change or be replaced, data persists and needs a consistent, centralized design.

**6 what is considered the central assest of the enterprise **

1 employees
2 technology
3 data
4 customers

The correct answer is:

3 — Data ✅

In modern enterprise architecture, data is considered the central asset because:

Applications, technology, and even business processes can change, but data has long-term value.
Data drives decision-making, analytics, compliance, and innovation.
Protecting and managing data well directly impacts the organization's competitiveness and resilience.

Employees, technology, and customers are all critical, but data is the persistent, reusable core asset across all business operations.

7 can unstructured data be as valuable as structured data

1 no
2 yes, but only in certain cases
3 yes always
4 it depends on the type of unstructured data

The correct answer is:

2 — Yes, but only in certain cases ✅

Unstructured data (like emails, documents, images, videos, social media posts, sensor readings) can be highly valuable—for example, in sentiment analysis, fraud detection, or medical imaging.

However:

Its value depends on context, quality, and the ability to process and interpret it.
Without the right tools and analysis, unstructured data might be hard to exploit effectively.

So, while it can match or exceed the value of structured data, it’s not always the case.

8 what does a data-centric enterprise have besides a tech implementation

1 an associated culture and mindset
2 no associated culture or mindset
3 only a tech implementation
4 only an associated culture and mindset

The correct answer is:

1 — An associated culture and mindset ✅

A data-centric enterprise isn’t just about having the right technology stack; it also requires:

A culture that treats data as a shared, strategic asset.
A mindset where teams prioritize data quality, governance, and reuse.
Policies and processes that ensure data is consistent, accessible, and trusted.

Without that cultural foundation, even the best technology won’t deliver the full benefits of being data-centric.

9 how are data products described?

1 based on a common taxonomy and ontology
2 based on individual preferences
3 based on the type of data product
4 based on stage of creation and evolution

The correct answer is:

1 — Based on a common taxonomy and ontology ✅

Data products should be described consistently so they can be easily understood, discovered, and integrated across the enterprise.

Common taxonomy → shared classification system for organizing data products.
Common ontology → shared definitions of terms, relationships, and meaning, ensuring everyone interprets data the same way.

Using individual preferences (option 2) or inconsistent criteria (options 3 or 4) would lead to confusion and make data harder to reuse.

10 what is expected of stakeholders in a system

1 they should not make data-driven decisions
2 they should make data-driven decisions during development
3 they should make data-driven decisions only during production
4 they will make data-driven decisions to validate ideas and reduce uncertainty. These may be automated

The correct answer is:

4 — They will make data-driven decisions to validate ideas and reduce uncertainty. These may be automated ✅

In a well-designed system:

Stakeholders rely on data to guide choices, not just intuition.
Decisions are made throughout the lifecycle — from early validation of ideas to ongoing optimization.
Automation can speed up decision-making and reduce human bias, especially for repetitive or real-time scenarios.

This approach ensures that decisions are evidence-based and uncertainty is minimized.

11 what is required for the information supplied to improve observability

1 it must be unintelligible and unrelated
2 it must be intelligible and unrelated
3 it must be unintelligible and relatable
4 it must be intelligible and relatable

The correct answer is:

4 — It must be intelligible and relatable ✅

For observability, the information you collect should:

Be intelligible → clear, understandable, and meaningful to the people or systems consuming it.
Be relatable → connected to the system’s context, so it helps explain behaviors, issues, and performance.

If data is unclear or irrelevant, it won’t help identify problems or improve system understanding.

12 what is expected of an enterprise regarding control plane information collection

1 collect as little information as possible
2 collect information only for specific use cases
3 collect as much info as possible and use it to dynamically improve resilience, reliability, and performance
4 collect info only for expected future use cases

The correct answer is:

3 — Collect as much info as possible and use it to dynamically improve resilience, reliability, and performance ✅

In an enterprise:

The control plane monitors and manages systems, applications, and infrastructure.
Collecting comprehensive information allows the enterprise to detect issues early, optimize performance, and adapt dynamically.
Limiting data collection (options 1, 2, or 4) risks blind spots and reduces the system’s ability to self-heal or improve.

13 what is required for backing service

1 they cannot be exchanged during other operation
2 they can only be exchanged during deployment
3 they can be exchanged during operation without redeploying the application
4 they cannot be exchanged at all

The correct answer is:

3 — They can be exchanged during operation without redeploying the application ✅

A backing service (like databases, message queues, or caches) should be:

Replaceable at runtime without requiring changes or redeployment of the application.
This allows for flexibility, scaling, and resilience, letting the system swap services if needed (e.g., for upgrades, failover, or switching vendors).

Options 1, 2, and 4 would limit operational flexibility and violate the principles of modern, cloud-native design.

14 what is the requirement for services specific to an application?

1 they should be deployed and enabled as part of deploying an application
2 they should not be part of deploying an application
3 they should be specific to each environment
4 they should not be enabled during deployment

The correct answer is:

1 — They should be deployed and enabled as part of deploying an application ✅

Services that are specific to an application (like certain APIs, caches, or supporting microservices) must:

Be deployed together with the application to ensure it functions correctly.
Be enabled automatically during deployment to avoid manual configuration errors.

Options 2, 3, and 4 would risk the application failing to run properly or inconsistencies across environments.

15 which of following is true about a codebase

1 it can be forked for different deployment
2 different deployment may have different codebase
3 it must be singular even it different have configuration that apply to them
4 env specific code be included in the codebase

3 — It must be singular even if different deployments have configurations that apply to them ✅

In modern development practices:

A single codebase is maintained for all deployments to ensure consistency, easier maintenance, and streamlined updates.
Deployment-specific variations (like environment settings, feature flags, or secrets) are handled through configuration, not by duplicating or forking the code.

Options 1, 2, and 4 go against best practices because forking or including environment-specific code leads to complexity, drift, and maintenance challenges.

16 what is the importance of maintaining links in a codebase

1 it allows for forking of the codebase
2 it ensures that different deployment have different codebases
3 it defines the codebase and ensures it remains singular
4 it allows for env specific code to be included in the codebase

The correct answer is:

3 — It defines the codebase and ensures it remains singular ✅

Maintaining links (like references to modules, libraries, or dependencies) is important because:

It keeps the codebase coherent and unified.
Ensures all deployments use the same underlying code while allowing configuration to handle environment-specific differences.
Prevents fragmentation or duplication that can lead to maintenance challenges and inconsistencies.

Options 1, 2, and 4 would undermine the principle of a single, maintainable codebase.

17 what does the enterprise do with collected information

1 it use to maintain it's current security posture
2 it use to improve its security posture dynamically
3 it uses it to improve its security posture statically

The correct answer is:

2 — It uses it to improve its security posture dynamically ✅

In a modern enterprise:

Collected information (logs, metrics, alerts, access patterns) is continuously analyzed.
This allows the enterprise to adapt and enhance security in real time, responding to threats, vulnerabilities, or anomalous behavior.
Static approaches (option 3) or using data only for maintenance (option 1) are insufficient in dynamic, evolving environments.

18 how is access to individual enterprise resources granted

1 on a atomic basis
2 on a large basis
3 on a case-by-case basis
4 on a device-by-device basis

The correct answer is:

1 — On an atomic basis ✅

In a data-centric enterprise, access control is fine-grained:

Permissions are granted at the smallest logical unit of a resource (atomic level), such as a specific record, file, or service operation.
This ensures least-privilege access, improving security and reducing the risk of unauthorized use.

Options 2, 3, and 4 are less precise and do not provide the same level of control and protection.

19 what is an attribute of an independent service

1 being stateful
2 being able to share state with other services
3 being able to predict its halting externally
4 being stateless

The correct answer is:

4 — Being stateless ✅

An independent service in modern architectures (like microservices) is typically:

Stateless, meaning it does not retain client-specific data between requests.
This enables scalability, resilience, and easier deployment, because any instance of the service can handle any request.

Options 1 and 2 (stateful or sharing state) reduce independence, and option 3 (predicting halting externally) is not a defining attribute of independence.

20 what is required for publishing events

1 must have a defined schema and be published using registered topic
2 must have a defined schema and published using unregistered topic
3 they do not have defined schema
4 they can be published without using topic

The correct answer is:

1 — Must have a defined schema and be published using a registered topic ✅

For event-driven systems:

A defined schema ensures that consumers understand the structure and meaning of the event data.
Using a registered topic ensures discoverability, governance, and proper routing of events across the system.

Options 2, 3, and 4 would lead to ambiguity, integration issues, and potential system errors.

21 what must be clear when cloud services scale with demand

1 the cost of scaling
2 the root cause of the demand
3 the number of users
4 the type of service

The correct answer is:

1 — The cost of scaling ✅

When cloud services scale dynamically with demand, it’s essential to understand how much it will cost to handle increased load.

This allows the organization to budget appropriately and avoid unexpected expenses.
While understanding the number of users, root cause of demand, and type of service can be useful, the primary concern for scaling decisions is the financial impact.

22 how are cloud services charged

1 to the users of the service
2 to the developer of service
3 to the owner of service
4 to the admin of service

The correct answer is:

3 — To the owner of the service ✅

In cloud computing:

The service owner is responsible for costs associated with running the service, including compute, storage, networking, and any managed services used.
Users of the service typically consume the service but do not directly pay the underlying cloud provider unless a billing model is set up.
Developers or admins may influence costs but are not typically charged directly.