MuleSoft-Platform-Architect-I Salesforce Certified MuleSoft Platform Architect 1 Exam (WI25) Questions and Answers

Explanation

Correct Answer: Implement required changes to the Process API implementation so that, whenever possible, the Process API’s RAML definition remains unchanged.

*****************************************

Key requirement in the question is:

>> Approach that minimizes the impact of this change on API clients

Based on above:

>> Updating the RAML definition would possibly impact the API clients if the changes require any thing mandatory from client side. So, one should try to avoid doing that until really necessary.

>> Implementing the changes as a completely different API and then redirectly the clients with 3xx status code is really upsetting design and heavily impacts the API clients.

>> Organisations and IT cannot simply postpone the changes required until all API consumers acknowledge they are ready to migrate to a new Process API or API version. This is unrealistic and not possible.

The best way to handle the changes always is to implement required changes to the API implementations so that, whenever possible, the API’s RAML definition remains unchanged.

An Anypoint Virtual Private Cloud (VPC) provides a secure and private networking environment for MuleSoft applications, using specific architectural elements:

CIDR Blocks:

Anypoint VPCs utilize CIDR blocks to define IP ranges, allowing organizations to control and segment the VPC’s IP address space.

Port 6091 for HTTP Requests:

By default, HTTP requests can be made to workers on port 6091 from the public internet, providing an accessible entry point unless additional restrictions are applied.

Explanation of Correct Answers (B, E):

CIDR blocks enable IP range management for VPCs, and port 6091 is the default public entry port, which is part of Anypoint VPC’s default settings.

Explanation of Incorrect Options:

Option A (Ports 8081 and 8082) is incorrect; these are not default public ports for Anypoint VPC.

Option C (responsible for load balancing) is incorrect as load balancing requires a separate Dedicated Load Balancer (DLB).

Option D (round-robin load balancing) applies to DLBs, not directly to VPCs.

ReferencesFor more on VPC setup and networking, refer to MuleSoft documentation on VPC configurations and default port settings​.

Explanation

Correct Answer: They provide an additional layer of resilience on top of the underlying backend system, thereby insulating clients from extended failure of these systems.

*****************************************

In API-led connectivity,

>> Experience APIs - allow for innovation at the user interface level by consuming the underlying assets without being aware of how data is being extracted from backend systems.

>> Process APIs - compose data from various sources and combine them with orchestration logic to create higher level value

>> System APIs - reduce the dependency on the underlying backend systems by helping unlock data from backend systems in a reusable and consumable way.

However, they NEVER promise that they provide an additional layer of resilience on top of the underlying backend system, thereby insulating clients from extended failure of these systems.

https://dzone.com/articles/api-led-connectivity-with-mule

Scenario Analysis:

The API implementation is experiencing high CPU usage (over 90%) during bursts of requests, which correlates with slow response times, as indicated by a 99th percentile response time greater than 120 seconds.

The API implementation is running on Anypoint Runtime Fabric with two non-clustered replicas and has a reserved vCPU of 1.0 and a vCPU limit of 2.0.

The high CPU usage during bursts suggests that the current resources may not be sufficient to handle peak loads.

Evaluating the Options:

Option A (Correct Answer): Increasing the vCPU resources for each replica would provide more processing power to handle high traffic volumes, potentially reducing the response time during spikes. Since the CPU usage is consistently high during bursts, this option directly addresses the resource bottleneck.

Option B: Modifying the API client to split requests may reduce individual request load but could be complex to implement on the client side and may not fully address the high CPU issue.

Option C: Increasing the number of replicas could help distribute the load; however, with a high CPU load on each replica, adding more replicas without increasing CPU resources may not fully resolve the problem.

Option D: Throttling the client would reduce the number of requests, but this may not be acceptable if the client needs to maintain a high request rate. It also does not directly address the CPU limitations of the API implementation.

Conclusion:

Option A is the best choice as it addresses the root cause of high CPU usage by increasing the vCPU allocation, allowing the API to handle more requests efficiently. This should be pursued first before considering other options.

Refer to MuleSoft’s documentation on Runtime Fabric and vCPU resource allocation for more details on optimizing API performance in high-demand environments.

To achieve the goal of exposing financial data to a new mobile application while following MuleSoft’s best practices, the company should follow an API-led connectivity approach. This approach ensures minimal disruption to existing clients, maximizes reusability, and respects the separation of concerns across API layers.

Explanation of Solution:

Experience APIs for Client-Specific Requirements:

Create two new Experience APIs (EAPI-1 and EAPI-2) for the mobile application, tailored to meet the specific data and format requirements of the mobile application. These APIs encapsulate the client-specific needs and provide a custom interface without impacting other clients.

Process API Layer for Data Transformation:

By adding Mobile PAPI-2, we allow the mobile application to access the required subset of data, formatted according to the mobile application’s requirements. This approach ensures that data transformation and aggregation are handled in the Process layer, maintaining consistency and reusability across different applications.

Reuse of System APIs:

Both the new Mobile PAPI-2 and existing PAPI-1 access data from System APIs (SAPI-1 and SAPI-2), which continue to expose data from each legacy system in a consistent, reusable manner. This avoids duplicating logic and ensures that data access remains centralized and manageable.

Why Option A is Correct:

Option A aligns with MuleSoft’s best practices by isolating client-specific requirements in the Experience layer, utilizing Process APIs for data orchestration and transformation, and maintaining reusable System APIs for backend access.

This approach also ensures that the current API clients are not impacted, as new clients (e.g., the mobile app) interact with newly defined Experience APIs without modifying the existing API setup.

Explanation of Incorrect Options:

Option B: This option seems similar but lacks clarity on the separation of mobile-specific requirements and does not explicitly mention data transformation, which is essential in this scenario.

Option C: Creating a single mobile Experience API that exposes a subset of PAPI endpoints directly adds unnecessary complexity and may violate the separation of concerns, as transformation logic should not be in the Experience layer.

Option D: Deploying a new PAPI and using an API Proxy to redirect existing endpoints would add unnecessary complexity, disrupt the current API clients, and increase maintenance efforts.

ReferencesFor additional guidance, refer to MuleSoft documentation on API-led connectivity best practices and best practices for structuring Experience, Process, and System APIs​​.

Understanding the Data Flow Between Mule Runtimes and Control Plane:

When Mule applications are deployed on customer-hosted Mule runtimes, the MuleSoft-hosted control plane (Anypoint Platform) can monitor and manage these applications. However, due to data privacy and security, the control plane only collects specific types of information.

Typically, only metadata about the request and response (such as headers, status codes, and timestamps) is sent to the MuleSoft-hosted control plane. The actual payload data is not transmitted unless explicitly configured, ensuring that sensitive data remains within the customer's network.

Evaluating the Options:

Option A (Only the data): This is incorrect because the payload data itself is not automatically sent to the control plane in default configurations.

Option B (No data): This is incorrect as well; while the payload is not sent, metadata is still collected and sent to the control plane.

Option C (The data and metadata): This option is incorrect because data (payload) is not transmitted to the control plane by default.

Option D (Correct Answer): Only the metadata is sent to the MuleSoft-hosted control plane by default, aligning with MuleSoft’s design to prioritize security and data privacy for customer-hosted runtimes.

Conclusion:

Option D is the correct answer, as by default, only metadata is sent to the MuleSoft-hosted control plane, and not the payload. This configuration is designed to protect sensitive data from being exposed outside the customer’s hosted environment.

For more details, refer to MuleSoft’s documentation on telemetry data collected in customer-hosted Mule runtimes and the MuleSoft control plane.

Understanding Control and Runtime Planes:

Control Plane: The control plane is responsible for managing, monitoring, and deploying Mule applications. In a Private Cloud Edition (PCE), this control plane is deployed on-premises within the customer’s infrastructure, meeting data residency and security requirements.

Runtime Plane: The runtime plane consists of Mule runtimes that execute Mule applications. By hosting these runtimes within the customer’s infrastructure, data and metadata can remain local, which complies with corporate policies regarding data residency.

Evaluating the Options:

Option A: Using Anypoint Platform Private Cloud Edition for the control plane and the MuleSoft-hosted runtime plane would not meet the requirement, as the runtime plane is hosted by MuleSoft and would not keep data local.

Option B: The MuleSoft-hosted control plane with Anypoint Runtime Fabric for the runtime plane would still mean that metadata is managed in MuleSoft’s cloud, which does not comply with the requirement to keep data and metadata on-premises.

Option C: A MuleSoft-hosted control plane and customer-hosted Mule runtimes also mean that metadata resides in the cloud, not on-premises, failing the residency requirement.

Option D (Correct Answer): Anypoint Platform Private Cloud Edition (PCE) for the control plane and customer-hosted Mule runtimes fulfill both requirements, as both the control plane and runtime plane would be hosted within the customer’s data center.

Conclusion:

Option D is the correct answer, as it ensures that both the control plane and runtime plane are hosted on-premises, allowing data and metadata to reside locally per the corporate IT policy.

Refer to MuleSoft's documentation on Private Cloud Edition deployment and on-premise runtime configurations for further details.

In MuleSoft CloudHub, autoscaling is essential to managing application load efficiently. CloudHub supports horizontal scaling based on CPU usage, which is well-suited to applications experiencing variable demand and needing responsive resource allocation.

Autoscaling on CloudHub:

Horizontal scaling increases the number of workers in response to CPU usage thresholds, allowing the application to handle higher loads dynamically. This approach improves performance without downtime or manual intervention.

Why Option C is Correct:

Setting up autoscaling based on CPU usage aligns with MuleSoft’s best practices for scalable and responsive applications on CloudHub, particularly in an environment with fluctuating load patterns.

Option C correctly leverages CloudHub’s autoscaling features based on resource metrics, which are part of CloudHub’s managed scaling solutions.

Explanation of Incorrect Options:

Option A (based on HTTP request thresholds) and Option B (separate policies for CPU and memory) do not represent CloudHub’s recommended scaling practices.

Option D suggests vertical scaling based on response time, which is not how CloudHub handles autoscaling.

ReferencesFor more on CloudHub’s autoscaling configuration, refer to MuleSoft documentation on CloudHub autoscaling policies​.

Explanation

Correct Answer: API Consumer requests access to API >> API Client implementes logic to call an API >> API routes the request to >> API Implementation

*****************************************

>> API consumer does not implement any logic to invoke APIs. It is just a role. So, the option stating "API Consumer implementes logic to call an API" is INVALID.

>> API Implementation does not route any requests. It is a final piece of logic where functionality of target systems is exposed. So, the requests should be routed to the API implementation by some other entity. So, the options stating "API Implementation routes the request to >> API" is INVALID

>> The statements in one of the options are correct but sequence is wrong. The sequence is given as "API Client implementes logic to call an API >> API Consumer requests access to API >> API routes the request to >> API Implementation". Here, the statements in the options are VALID but sequence is WRONG.

>> Right option and sequence is the one where API consumer first requests access to API on Anypoint Exchange and obtains client credentials. API client then writes logic to call an API by using the access client credentials requested by API consumer and the requests will be routed to API implementation via the API which is managed by API Manager.

Explanation

Correct Answer: Data on past API invocations to help identify anomalies and usage patterns across various APIs

*****************************************

API Invocation metrics provided by Anypoint Platform:

>> Does NOT provide any Return Of Investment (ROI) related information. So the option suggesting it is OUT.

>> Does NOT provide any information w.r.t how APIs are reused, whether there is effective usage of APIs or not etc...

>> Does NOT prodive any prediction information as such to help us proactively identify any future policy violations.

So, the kind of data/information we can get from such metrics is on past API invocations to help identify anomalies and usage patterns across various APIs.

In the API-led connectivity approach, APIs are categorized into Experience, Process, and System layers:

Enterprise Data Model Scope:

The Enterprise Data Model (EDM) generally supports System APIs and some Process APIs by defining standard data structures used across the organization. Experience APIs, however, are tailored to specific applications or interfaces and are less likely to be served directly by the EDM, as they may require customized data representations to meet the unique needs of each user interface.

Why Option C is Correct:

The Mobile Account Management EAPI serves mobile-specific needs and often requires data formatted differently from the standardized data models. Thus, it would be outside the direct scope of the EDM and might employ custom mappings to fit mobile application requirements.

Explanation of Incorrect Options:

Option A (Customer SAPI), Option B (Customer Lookup PAPI), and Option D (Service SAPI) would typically align with the EDM as they are closer to the core data and services the EDM supports.

ReferencesFor additional guidance, review MuleSoft's best practices on API-led connectivity and data modeling​.

Explanation

Correct Answer: The client ID and secret obtained from Anypoint Exchange for the API instance in the STAGING environment

*****************************************

>> We CANNOT use the client ID and secret of Anypoint Platform account or any individual environments for accessing the APIs

>> As the type of policy that is enforced on the API in question is "Client ID Enforcment Policy", OAuth token based access won't work.

Right way to access the API is to use the client ID and secret obtained from Anypoint Exchange for the API instance in a particular environment we want to work on.

References:

Managing API instance Contracts on API Manager

https://docs.mulesoft.com/api-manager/1.x/request-access-to-api-task

https://docs.mulesoft.com/exchange/to-request-access

https://docs.mulesoft.com/api-manager/2.x/policy-mule3-clie nt-id-based-policies

Explanation

Correct Answer: Search Anypoint exchange for a suitable existing fallback API, and then implement invocations to this fallback API in addition to the order API

*****************************************

>> It is not ideal and good approach, until unless there is a pre-approved agreement with the API clients that they will receive a HTTP 3xx temporary redirect status code and they have to implement fallback logic their side to call another API.

>> Creating separate entry of same Order API in API manager would just create an another instance of it on top of same API implementation. So, it does NO GOOD by using clone od same API as a fallback API. Fallback API should be ideally a different API implementation that is not same as primary one.

>> There is NO option currently provided by Anypoint HTTP Connector that allows us to invoke a fallback API when we receive certain HTTP status codes in response.

The only statement TRUE in the given options is to Search Anypoint exchange for a suitable existing fallback API, and then implement invocations to this fallback API in addition to the order API.

When an API implementation is deployed on CloudHub without configuring automatic restarts in case of worker non-responsiveness, MuleSoft’s monitoring and alerting behavior is as follows:

API Invocation Alerts:

If no clients are invoking the API, there will be no invocation alerts triggered, as alerts related to invocations depend on actual client requests.

Implementation-Level Alerts:

Even without invocation, CloudHub can still monitor the state of the API implementation. If the worker becomes unresponsive, an alert related to the API implementation's health or availability could still be raised.

Why Option C is Correct:

This option correctly identifies that no invocation-related alerts would be triggered in the absence of client requests, while implementation-level alerts could still be generated based on the worker's state.

ReferencesFor additional information, check MuleSoft documentation on CloudHub monitoring and alert configurations to understand worker status alerts versus invocation alerts​.

Scenario Analysis:

The eCommerce company’s Product Details feature requires low response time and consistent data for a feature where data rarely changes (only once or twice a year).

The database response time is slow due to high volume, so querying the database directly on each request would lead to poor performance and higher response times.

Optimal Solution Requirements:

Low Response Time: Data retrieval should be fast and not depend on database performance.

Fault Tolerance and Data Consistency: Cached or stored data should be consistent and resilient in case of database unavailability, as the product details data changes infrequently.

Evaluating the Options:

Option A: Using a Cache scope would temporarily store the product details in memory, which could improve performance but might not be suitable for infrequent updates (only twice a year), as cache expiration policies typically require shorter durations.

Option B: Storing product details in Anypoint MQ and then retrieving it through a subscriber is not suitable for this use case. Anypoint MQ is better for messaging rather than as a data storage mechanism.

Option C (Correct Answer): Using an object store to store and retrieve product details is ideal. Object stores in MuleSoft are designed for persistent storage of key-value pairs, which allows storing data retrieved from the database initially. This provides quick, consistent access without querying the database on every request, aligning with requirements for low response time, fault tolerance, and data consistency.

Option D: Selecting data directly from the database for each request would not meet the performance requirement due to known slow response times from the database.

Conclusion:

Option C is the best answer, as using an object store allows caching the infrequently updated product details. This approach reduces the dependency on the database, significantly improving response time and ensuring consistent data.

Refer to MuleSoft documentation on Object Store v2 and best practices for data caching to implement this solution effectively.

Understanding API Functional Monitoring:

API Functional Monitoring is a feature within MuleSoft's Anypoint Platform that enables users to monitor the health and performance of APIs and endpoints by running functional tests at scheduled intervals.

It checks whether APIs are functioning as expected by running test calls and then evaluating if the response meets the desired conditions. This is particularly useful for testing endpoint availability, checking for specific data in responses, and measuring API performance over time.

Component Features:

Scheduled Intervals: Functional monitoring allows configuring tests to run at regular intervals, such as every minute, hour, or day, depending on the monitoring requirements.

Reports on Test Pass/Fail Status: After each test run, API Functional Monitoring reports whether the API passed or failed the test conditions.

Performance Statistics: It displays metrics like average response time, success rate, and error rates, giving insights into API health and performance.

Evaluating the Options:

Option A (API Analytics): API Analytics provides insights on API usage and metrics but does not involve scheduled tests for pass/fail status or endpoint health checks.

Option B (Anypoint Monitoring Dashboards): These dashboards display API metrics but do not actively test API endpoints or provide pass/fail reporting on a scheduled basis.

Option C (Correct Answer): API Functional Monitoring fits the description, as it is designed to monitor API and endpoint health with scheduled test runs and display statistics about performance.

Option D (Anypoint Runtime Manager Alerts): Runtime Manager alerts notify users of issues with application status but do not actively test endpoints at scheduled intervals.

Conclusion:

Option C (API Functional Monitoring) is the correct answer because it provides the necessary tools to test API functionality, monitor endpoint health, and display performance statistics in real-time.

Refer to MuleSoft documentation on API Functional Monitoring for further guidance on setting up and configuring these tests in Anypoint Platform.

Understanding API Autodiscovery in MuleSoft:

API Autodiscovery links an API implementation in Anypoint Platform with its configuration in API Manager. This is controlled by the API ID which is set in the API Autodiscovery element in the Mule application.

The API ID remains consistent across minor updates to the API asset version in Exchange (e.g., from 2.0.0 to 2.0.1) as long as it is the same API.

Effect of Asset Version Update on API Autodiscovery:

When the asset version is updated (e.g., from 2.0.0 to 2.0.1), the API ID remains the same. Therefore, no changes are needed in the Autodiscovery configuration within the Mule application. The Autodiscovery will continue to link the API implementation to the latest version in API Manager.

Evaluating the Options:

Option A: Incorrect, as the API ID does not automatically change with minor asset version updates.

Option B: Incorrect, as the API ID remains the same, so no update is needed in the API implementation.

Option C (Correct Answer): The API ID does not change, so no changes are necessary in the API implementation for the new asset version.

Option D: Incorrect, as there is no need to update the API implementation in the Autodiscovery global element for minor version changes.

Conclusion:

Option C is the correct answer, as the API ID remains unchanged with minor version updates, and no changes are needed in the API Autodiscovery configuration.

Refer to MuleSoft documentation on API Autodiscovery and version management for more details.

Explanation

Correct Answer: Manually provisioned customer-hosted runtime plane and customer-hosted control plane

*****************************************

There are two key factors that are to be taken into consideration from the scenario given in the question.

>> Company requires both data and metadata to be resided within the corporate firewall

>> Company would like to go with customer-hosted infrastructure.

Any deployment model that is to deal with the cloud directly or indirectly (Mulesoft-hosted or Customer's own cloud like Azure, AWS) will have to share atleast the metadata.

Application data can be controlled inside firewall by having Mule Runtimes on customer hosted runtime plane. But if we go with Mulsoft-hosted/ Cloud-based control plane, the control plane required atleast some minimum level of metadata to be sent outside the corporate firewall.

As the customer requirement is pretty clear about the data and metadata both to be within the corporate firewall, even though customer wants to move to production as quickly as possible, unfortunately due to the nature of their security requirements, they have no other option but to go with manually provisioned customer-hosted runtime plane and customer-hosted control plane.

In the initial meeting for establishing a Center for Enablement (C4E), it’s essential to lay the foundational vision, objectives, and guiding principles for the team. Here’s why this is crucial:

Clear Vision and Mission:

Defining the mission statement and objectives at the start ensures alignment within the organization and clarifies the C4E’s role in supporting API-led development and integration practices.

Guiding Principles:

Establishing guiding principles will help the C4E maintain consistent practices and strategies across projects. This serves as a framework for decisions and fosters shared understanding among IT leaders and stakeholders.

Explanation of Correct Answer (A):

By prioritizing the C4E’s objectives and mission, the organization builds a solid foundation, paving the way for subsequent meetings focused on technical standards, processes, and operating models.

Explanation of Incorrect Options:

Option B (API monetization) and Option C (common services best practices) are specific topics better suited for later discussions.

Option D (specifying the operating model) is an important step but typically follows the establishment of the C4E’s objectives and vision.

ReferencesFor more on C4E objectives and foundational setup, refer to MuleSoft’s documentation on establishing a C4E and the roles and mission statements recommended for such initiatives​.

*****************************************

>> Application Network is a disposable architecture.

>> Which means, it can be altered without disturbing entire architecture and its components.

>> It bends as per requirements or design changes but does not break

Explanation

Correct Answer: Create separate Anypoint VPCs for non-production and production environments, then configure connections to the backend systems in the corresponding customer-hosted environments.

*****************************************

>> Creating different Business Groups does NOT make any difference w.r.t accessing the non-prod and prod customer-hosted environments. Still they will be accessing from both Business Groups unless process network restrictions are put in place.

>> We need to modify or couple the Mule Application Implementations with the environment. In fact, we should never implements application coupled with environments by binding them in the properties. Only basic things like endpoint URL etc should be bundled in properties but not environment level access restrictions.

>> IP addresses on CloudHub are dynamic until unless a special static addresses are assigned. So it is not possible to setup firewall rules in customer-hosted infrastrcture. More over, even if static IP addresses are assigned, there could be 100s of applications running on cloudhub and setting up rules for all of them would be a hectic task, non-maintainable and definitely got a good practice.

>> The best practice recommended by Mulesoft (In fact any cloud provider), is to have your Anypoint VPCs seperated for Prod and Non-Prod and perform the VPC peering or VPN tunneling for these Anypoint VPCs to respective Prod and Non-Prod customer-hosted environment networks.

Explanation

Correct Answer: They are typically written using MUnit to run in an embedded Mule runtime that does not require external connectivity.

*****************************************

Below TWO are characteristics of Integration Tests but NOT unit tests:

>> They test the validity of input and output of source and target systems.

>> They must be triggered by an external client tool or event source.

It is NOT TRUE that Unit Tests must be run in a unit testing environment with dedicated Mule runtimes for the environment.

MuleSoft offers MUnit for writing Unit Tests and they run in an embedded Mule Runtime without needing any separate/ dedicated Runtimes to execute them. They also do NOT need any external connectivity as MUnit supports mocking via stubs.

https://dzone.com/articles/munit-framework

Explanation

Correct Answer: Use a horizontal CloudHub autoscaling policy that triggers on CPU utilization greater than 70%

*****************************************

The scenario in the question is very clearly stating that the usual traffic in the year is pretty well handled by the existing worker configuration with CPU running well below 70%. The problem occurs only "sometimes" occasionally when there is spike in the number of orders coming in.

So, based on above, We neither need to permanently increase the size of each worker nor need to permanently increase the number of workers. This is unnecessary as other than those "occasional" times the resources are idle and wasted.

We have two options left now. Either to use horizontal Cloudhub autoscaling policy to automatically increase the number of workers or to use vertical Cloudhub autoscaling policy to automatically increase the vCore size of each worker.

Here, we need to take two things into consideration:

1. CPU

2. Order Submission Rate to JMS Queue

>> From CPU perspective, both the options (horizontal and vertical scaling) solves the issue. Both helps to bring down the usage below 90%.

>> However, If we go with Vertical Scaling, then from Order Submission Rate perspective, as the application is still being load balanced with two workers only, there may not be much improvement in the incoming request processing rate and order submission rate to JMS queue. The throughput would be same as before. Only CPU utilization comes down.

>> But, if we go with Horizontal Scaling, it will spawn new workers and adds extra hand to increase the throughput as more workers are being load balanced now. This way we can address both CPU and Order Submission rate.

Hence, Horizontal CloudHub Autoscaling policy is the right and best answer.

Default Alert Capabilities in Anypoint Platform:

Anypoint Platform provides out-of-the-box alerting capabilities for monitoring API invocation conditions, including setting thresholds for the number of invocations.

Alerts can be configured for conditions such as high or low traffic (invocations exceeding or falling below a defined threshold).

Evaluating the Options:

Option A: Anypoint Platform does not provide direct alerting based on DNS records.

Option B: Anypoint Platform does not provide default alerts based on whether invocations use TLS/SSL; this would require custom configuration.

Option C: Geolocation-based alerting is not natively supported in Anypoint Platform.

Option D (Correct Answer): Alerts based on API invocation thresholds (e.g., invocations falling below a set threshold) are supported and can be configured as part of the default Anypoint alerting functionality.

Conclusion:

Option D is correct, as Anypoint Platform allows configuring alerts based on the number of API invocations falling below or exceeding a threshold.

Refer to MuleSoft's documentation on Anypoint Monitoring and alert configurations for more details.

In API-led design, an Experience API is enhanced to improve how data is delivered to end-user applications. One primary reason to enhance an Experience API is when new data standards, such as a Canonical Data Model, are adopted. Here’s why:

Canonical Data Model (CDM):

Adopting a CDM standardizes data representations across the organization, making APIs more consistent and easier to consume across various services and applications.

Updating the Experience API ensures that it delivers data in this standardized format, improving interoperability and reusability.

Explanation of Correct Answer (D):

A CDM impacts the structure and types of data the API provides, and this update would be directly relevant to an Experience API, as it is the primary point of interaction for applications.

Explanation of Incorrect Options:

Option A involves adapting to consumer-specific standards, which is against API-led design principles.

Option B involves changes in System APIs, which don’t directly mandate changes to the Experience API unless data formatting adjustments are required.

Option C (IP Allowlist) relates to security rather than API design and would not motivate a functional enhancement of the API.

ReferencesFor more details on the use of Canonical Data Models in API-led architecture, refer to MuleSoft’s guidelines on data standardization and Experience API best practices​.

Explanation

Correct Answer: Ask the Marketing Department to interact with a mocking implementation of the API using the automatically generated API Console.

*****************************************

As per MuleSoft's IT Operating Model:

>> API consumers need NOT wait until the full API implementation is ready.

>> NO technical test-suites needs to be shared with end users to interact with APIs.

>> Anypoint Platform offers a mocking capability on all the published API specifications to Anypoint Exchange which also will be rich in documentation covering all details of API functionalities and working nature.

>> No needs of arranging days of workshops with end users for feedback.

API consumers can use Anypoint Exchange features on the platform and interact with the API using its mocking feature. The feedback can be shared quickly on the same to incorporate any changes.

Center for Enablement (C4E) KPIs:

A Center for Enablement (C4E) in MuleSoft focuses on enabling self-service and reuse by providing APIs that can be consumed across the organization. A key metric of success is how many consumers are utilizing the published APIs.

The number of consumers who have requested and received access to an API indicates the level of adoption and reuse, which aligns with the goals of a C4E.

Evaluating the Options:

Option A: This metric could indicate deployment automation, but it is not a direct measure of C4E’s success in enabling API reuse and consumption.

Option B: Bandwidth usage per API implementation provides insight into API traffic but does not measure C4E enablement or consumer engagement.

Option C: The number of developers downloading an API specification can be an indicator of interest but does not confirm actual usage or enablement.

Option D (Correct Answer): The number of consumers who have requested and received access to an API in production is a key metric indicating API adoption and reuse, which aligns with C4E’s goals.

Conclusion:

Option D is the correct answer as it provides a direct measure of consumer engagement and adoption, indicating the success of the C4E in promoting API usage across the organization.

Refer to MuleSoft’s documentation on C4E KPIs and API usage metrics for additional insights.

When defining data models for System APIs without an established Enterprise Data Model, MuleSoft recommends mirroring the back-end systems' data types to achieve quick and effective integration without adding complexity. This approach has several benefits:

Alignment with Backend Systems:

Mirroring data types ensures consistency with backend data sources, which simplifies integration, reduces mapping requirements, and minimizes potential data transformation issues.

Flexibility for Future Enhancements:

By retaining close alignment with backend data structures, System APIs can evolve to support an Enterprise Data Model in the future without immediate restructuring.

Explanation of Incorrect Options:

Option A (exposing misspellings) is not recommended as System APIs should still ensure a professional and coherent interface.

Option C (custom naming) complicates the API structure without adding immediate value in the absence of a clear data model.

Option D (exposing all fields) is unnecessary and can reduce performance and add complexity.

ReferencesRefer to MuleSoft best practices for data modeling in System APIs for additional information on mirroring backend systems​.

Explanation

Correct Answer: At the API proxy

*****************************************

>> API Policies can be enforced at two places in Mule platform.

>> One - As an Embedded Policy enforcement in the same Mule Runtime where API implementation is running.

>> Two - On an API Proxy sitting in front of the Mule Runtime where API implementation is running.

>> As the deployment scenario in the question has API Proxy involved, the policies will be enforced at the API Proxy.

Given that the application is designed to handle only one process at a time to maintain data consistency, here’s why horizontal scaling won’t increase the processing limit:

Single-Process Constraint:

The application limits to processing one transaction at a time due to its design for consistency, meaning horizontal scaling (adding more workers) does not increase processing speed beyond this limit.

Execution Time:

Since each request takes 200 ms, five requests per second is the maximum processing threshold. Increasing the number of workers does not bypass this single-process limitation.

Explanation of Correct Answer (A):

The scalability remains at five requests per second, as this constraint is intrinsic to the application’s design.

Explanation of Incorrect Options:

Option B suggests a change in execution time, which horizontal scaling does not affect.

Option C assumes doubling the throughput, which isn’t possible due to the single-threaded nature of the application.

Option D suggests horizontal scaling cannot apply, which is incorrect; however, scaling does not increase throughput in this context.

ReferencesFor more on understanding scaling and concurrency in Mule applications, see MuleSoft’s documentation on application performance and scaling limitations​.

Explanation

Correct Answer: Convince the development team of the product API to adopt the API data model of the Order API such that integration logic of the Order API can work with one consistent internal data model

*****************************************

Key details to note from the given scenario:

>> Power relationship between Order API and Product API is customer/supplier

So, as per below rules of "Power Relationships", the caller (in this case Order API) would request for features to the called (Product API team) and the Product API team would need to accomodate those requests.

Explanation

Correct Answer: The remaining capacity allowed by the API implementation.

*****************************************

>> Reference: https://docs.mulesoft.com/api-manager/2.x/rate-limiting-and-throttling-sla-based-policies#response-headers

Explanation

Correct Answer: It makes API Manager aware of API implementations and hence enables it to enforce policies.

*****************************************

>> API Autodiscovery is a mechanism that manages an API from API Manager by pairing the deployed application to an API created on the platform.

>> API Management includes tracking, enforcing policies if you apply any, and reporting API analytics.

>> Critical to the Autodiscovery process is identifying the API by providing the API name and version.

References:

https://docs.mulesoft.com/api-manager/2.x/api-auto-discovery-new-concept

https://docs.mulesoft.com/api-manager/1.x/api-auto-discovery

https://docs.mulesoft.com/api-manager/2.x/api-auto-discovery-new-conc ept

Explanation

Correct Answer: IP whitelist

*****************************************

>> OAuth 2.0 access token and Client ID enforcement policies are VERY common to apply on Experience APIs as API consumers need to register and access the APIs using one of these mechanisms

>> JSON threat protection is also VERY common policy to apply on Experience APIs to prevent bad or suspicious payloads hitting the API implementations.

>> IP whitelisting policy is usually very common in Process and System APIs to only whitelist the IP range inside the local VPC. But also applied occassionally on some experience APIs where the End User/ API Consumers are FIXED.

>> When we know the API consumers upfront who are going to access certain Experience APIs, then we can request for static IPs from such consumers and whitelist them to prevent anyone else hitting the API.

However, the experience API given in the question/ scenario is intended to work with a consumer mobile phone or tablet application. Which means, there is no way we can know all possible IPs that are to be whitelisted as mobile phones and tablets can so many in number and any device in the city/state/country/globe.

So, It is very LEAST LIKELY to apply IP Whitelisting on such Experience APIs whose consumers are typically Mobile Phones or Tablets.

Explanation

Correct Answer: Apply a basic authentication - LDAP policy; the internal Active Directory will be configured as the LDAP source for authenticating users.

*****************************************

>> IP Whitelisting does NOT fit for this purpose. Moreover, the users workstations may not necessarily have static IPs in the network.

>> OAuth 2.0 enforcement requires a client provider which isn't in the organizations system components.

>> It is not an effective approach to let every user create separate client credentials and configure those for their usage.

The effective way it to apply a basic authentication - LDAP policy and the internal Active Directory will be configured as the LDAP source for authenticating users.

Understanding Coarse-Grained vs. Fine-Grained API Design:

A coarse-grained design consolidates multiple operations within a single API, leading to fewer APIs but with more complex implementations. Conversely, a fine-grained design breaks down functionalities into smaller, more specific APIs, resulting in simpler implementations but a larger number of APIs.

Evaluating the Options:

Option A (Correct Answer): With a coarse-grained design, each API handles more functionalities, increasing the complexity of each API implementation as it needs to manage more use cases and logic.

Option B: A coarse-grained design typically reduces the number of APIs, so fewer discoverable assets are available.

Option C: Fewer APIs generally mean fewer connections between them in the application network.

Option D: Network infrastructure usage may actually decrease with fewer APIs, as there are fewer calls between APIs.

Conclusion:

Option A is the correct answer, as the complexity of each API implementation increases in a coarse-grained design due to the consolidation of multiple functionalities into single APIs.

Refer to MuleSoft’s documentation on API design principles and best practices for coarse-grained vs. fine-grained API implementation.

Explanation

Correct Answer: A higher number of discoverable API-related assets in the application network.

*****************************************

>> We do NOT get faster response times in fine-grained approach when compared to coarse-grained approach.

>> In fact, we get faster response times from a network having coarse-grained APIs compared to a network having fine-grained APIs model. The reasons are below.

Fine-grained approach:

1. will have more APIs compared to coarse-grained

2. So, more orchestration needs to be done to achieve a functionality in business process.

3. Which means, lots of API calls to be made. So, more connections will needs to be established. So, obviously more hops, more network i/o, more number of integration points compared to coarse-grained approach where fewer APIs with bulk functionality embedded in them.

4. That is why, because of all these extra hops and added latencies, fine-grained approach will have bit more response times compared to coarse-grained.

5. Not only added latencies and connections, there will be more resources used up in fine-grained approach due to more number of APIs.

That's why, fine-grained APIs are good in a way to expose more number of resuable assets in your network and make them discoverable. However, needs more maintenance, taking care of integration points, connections, resources with a little compromise w.r.t network hops and response times.

Processing Requirements and Storage Mechanism:

The Reservation Mule application will be deployed to multiple CloudHub workers/replicas, meaning that each worker must share state information to handle records exactly once. This requires a shared storage mechanism where state can be stored and accessed by multiple instances to avoid duplicate processing of the same records.

A Persistent Object Store in Anypoint Platform can be used to store records in a way that is accessible across multiple workers, providing a reliable mechanism for "exactly once" processing.

Evaluating the Options:

Option A (Correct Answer): A Persistent Object Store is designed to retain data across different application instances and can be shared by all workers on CloudHub. It helps achieve idempotency by ensuring that a record is processed exactly once.

Option B: Runtime Fabric Object Store is used for applications deployed in Anypoint Runtime Fabric, not CloudHub. This option would not be compatible with the CloudHub deployment.

Option C: A Non-persistent Object Store does not retain data across application restarts or different instances, making it unsuitable for the requirement of synchronized storage for exactly-once processing.

Option D: An In-memory Mule Object Store is local to each worker and is not shared across instances, so it does not meet the requirement for a shared storage mechanism accessible to all CloudHub workers.

Conclusion:

Option A is the correct answer, as a Persistent Object Store allows data sharing across multiple CloudHub workers, enabling them to synchronize and achieve "exactly once" processing of passenger records with minimal development effort.

Refer to MuleSoft’s documentation on Object Store configurations and usage for best practices on handling state across distributed instances.

Explanation

Correct Answer: When server-side load-balanced TLS mutual authentication is required between API implementations and API clients

*****************************************

Fact/ Memory Tip: Although there are many benefits of CloudHub Dedicated Load balancer, TWO important things that should come to ones mind for considering it are:

>> Having URL endpoints with Custom DNS names on CloudHub deployed apps

>> Configuring custom certificates for both HTTPS and Two-way (Mutual) authentication.

Coming to the options provided for this question:

>> We CANNOT use DLB to perform cross-region load balancing between separate deployments of the same Mule application.

>> We can have mapping rules to have more than one DLB URL pointing to same Mule app. But vicevera (More than one Mule app having same DLB URL) is NOT POSSIBLE

>> It is true that DLB helps to setup custom DNS names for Cloudhub deployed Mule apps but NOT true for apps deployed to Customer-hosted Mule Runtimes.

>> It is true to that we can load balance API invocations across multiple CloudHub workers using DLB but it is NOT A MUST. We can achieve the same (load balancing) using SLB (Shared Load Balancer) too. We DO NOT necessarily require DLB for achieve it.

So the only right option that fits the scenario and requires us to use DLB is when TLS mutual authentication is required between API implementations and API clients.

Explanation

Correct Answer: Unit tests (Whitebox)

*****************************************

Understanding JSON Threat Protection Policies:

JSON Threat Protection policies are used to protect APIs from attacks that exploit JSON payloads, such as oversized payloads, deeply nested objects, and excessive array elements. This helps prevent Denial of Service (DoS) attacks and other malicious payload-related threats.

These policies are typically applied to safeguard APIs that are directly exposed to external clients, where the risk of receiving malicious payloads is highest.

API-led Connectivity Layers:

Experience Layer: This layer is designed to expose APIs to end-users or external API clients, often acting as the interface that interacts with users or applications.

Process Layer: This layer is used for orchestration and aggregation of data from various System APIs, typically operating within a trusted environment and not directly exposed to external clients.

System Layer: This layer provides access to backend systems and databases, often within the organization’s secure environment and not directly accessible to external clients.

Evaluating the Options:

Option A (All layers): While JSON Threat Protection can technically be applied to all layers, it is most commonly applied at the Experience layer, where APIs are exposed to external traffic and are more vulnerable to attacks.

Option B (System layer): The System layer is generally not exposed to external clients directly, so JSON Threat Protection is less critical here.

Option C (Process layer): Similar to the System layer, the Process layer is typically internal and not exposed directly to external clients, so JSON Threat Protection is less commonly applied.

Option D (Correct Answer): The Experience layer is the correct answer because it is the layer that directly interacts with external clients, making it the primary target for malicious payloads. Applying JSON Threat Protection here effectively protects the application network from external threats.

Conclusion:

Option D is the correct answer, as the Experience layer is the most common layer for applying JSON Threat Protection policies to protect against external attacks.

For further reference, consult MuleSoft's documentation on API security policies and best practices for securing APIs at the Experience layer.