Pass the Salesforce Salesforce MuleSoft MuleSoft-Platform-Architect-I Questions and answers with CertsForce

To monitor an application network and predict issues without custom scripts, policy violation metrics are critical. They provide insights into potential problems by tracking instances where API usage does not conform to defined policies. Here’s why this approach is suitable:

Predictive Monitoring:

Tracking API policy violations (such as rate limits or spike controls being hit) can indicate surges in traffic or misuse, which may lead to throttling or service degradation if not addressed.

By monitoring these violations, teams can proactively adjust limits or optimize API handling to prevent actual failures.

No Custom Scripting Needed:

Policy violation metrics are available within MuleSoft’s Anypoint Monitoring, meaning there’s no need to implement custom solutions or external tools to gather and interpret this data.

Explanation of Incorrect Options:

Option B (effectiveness based on reuse) does not directly predict failures.

Option C (past invocation counts) offers historical usage data but does not inherently identify issues.

Option D (ROI from API invocation) is a business metric and does not provide technical insights for failure prediction.

ReferencesFor more details on leveraging policy violation metrics for proactive monitoring, refer to MuleSoft documentation on Anypoint Monitoring​​.

Explanation

Correct Answer: Create a bounded-context model for the system layer to closely match the backend data model, and add an anti-corruption layer to let the different bounded contexts cooperate across the system and process layers

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

>> Canonical models are not an option here as the organization has already put in efforts and created bounded-context models for Experience and Process APIs.

>> Anti-corruption layers for ALL APIs is unnecessary and invalid because it is mentioned that experience and process APIs share same bounded-context model. It is just the System layer APIs that need to choose their approach now.

>> So, having an anti-corruption layer just between the process and system layers will work well. Also to speed up the approach, system APIs can mimic the backend system data model.

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.

Explanation

Correct Answer: Guarding against Denial of Service attacks

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

>> Backend system overloading can be handled by enforcing "Spike Control Policy"

>> Logging HTTP requests and responses can be done by enforcing "Message Logging Policy"

>> Credentials can be tamper-proofed using "Security" and "Compliance" Policies

However, unfortunately, there is no proper way currently on Anypoint Platform to guard against DOS attacks.

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.

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.

Explanation

Correct Answer: Create reusable assets, Make them discoverable so that LOB teams can self-serve and browse the APIs, Get active feedback and usage metrics.

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

Diagram, arrow Description automatically generated

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 Requirements:

The business needs to transfer daily data changes from the Salesforce Account and Case objects to AWS DynamoDB in a single JSON format.

Only a subset of attributes (5 from Account and 8 from Case) is required, so it is not necessary to include all 219 attributes of the Account object.

Design Approach:

A System API (SAPI) should be created for each Salesforce object (Account and Case), exposing only the required fields (5 attributes for Account and 8 for Case).

A Process API (PAPI) can be used to aggregate and transform the data from these SAPIs, combining the 13 selected attributes from Account and Case into a single JSON structure for DynamoDB.

Evaluating the Options:

Option A: Mimicking all attributes in the SAPI is inefficient and unnecessary, as only 13 attributes are required.

Option B: Replicating all attributes in DynamoDB is excessive and would result in higher storage and processing costs, which is unnecessary given the requirement for only a subset of attributes.

Option C: Implementing an Enterprise Data Model could be useful in broader data management but is not required here, as the focus is on a lightweight integration.

Option D (Correct Answer): Creating separate entities in SAPI for Account and Case with only the required attributes and using the PAPI to aggregate them into a single JSON is the most efficient and meets the requirements effectively.

Conclusion:

Option D is the best choice as it provides a lightweight, efficient design that meets the requirements by transferring only the necessary attributes and minimizing resource use.

Refer to MuleSoft's best practices for API-led connectivity and data modeling to structure SAPIs and PAPIs efficiently.