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

CloudHub Object Store v2 is a managed key-value store provided by MuleSoft to support various use cases where temporary data storage is required. Here’s why Option D is correct:

Key Length Support: Object Store v2 allows storage of keys with a length of up to 300 characters, making it suitable for applications needing flexible and descriptive keys.

Limitations on Size:

Object Store v2 is not intended for large payload storage and has a recommended size limit below 10 MB for each value. Payloads exceeding 15 MB may cause performance issues and are better suited to a file storage system or database.

Option B is incorrect because storing payloads above 15 MB exceeds Object Store’s optimal usage specifications.

Key-Value Limits: Object Store v2 is designed for moderate, transient storage needs, and does not support unlimited storage. Thus, Option A is incorrect.

Backward Compatibility: Object Store v2 does not support Mule 4 applications running Object Store v1. Option C is incorrect as Object Store v1 and v2 are distinct.

ReferencesFor more on CloudHub Object Store v2, refer to MuleSoft documentation on Object Store limitations and configuration​.

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.

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.

Role of Process API in API-led Connectivity:

In MuleSoft’s API-led connectivity approach, a Process API is used to coordinate, aggregate, and orchestrate data from various System APIs. It is primarily responsible for implementing business logic that spans multiple backend systems or entities, transforming and combining data as needed to support business processes.

Process APIs are not directly exposed to end-user clients; rather, they work between System APIs and Experience APIs, providing business logic and orchestration capabilities.

Evaluating the Options:

Option A: Process APIs are not typically responsible for large data dumps or data synchronization. That function would be handled by a System API or a specialized batch process.

Option B: Managing direct, secure communication between back-end systems and end-user clients is typically the role of Experience APIs rather than Process APIs.

Option C (Correct Answer): Process APIs are designed to coordinate and orchestrate calls to multiple other APIs in the network, which supports the automation of business processes.

Option D: Securing and managing communication with end-user clients is typically the role of Experience APIs, not Process APIs.

Conclusion:

Option C is the correct answer, as the main role of a Process API is to coordinate and orchestrate interactions between other APIs, enabling business processes to function seamlessly across multiple systems.

Refer to MuleSoft's API-led connectivity documentation for further explanation of the roles and responsibilities of Process APIs in an application network.

Explanation

Correct Answer: Allow System APIs to return data that is NOT currently required by the identified Process or Experience APIs.

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

>> All customizations for the end-user application should be handled in "Experience API" only. Not in Process API

>> We should use tiered approach but NOT always by creating exactly one API for each of the 3 layers. Experience APIs might be one but Process APIs and System APIs are often more than one. System APIs for sure will be more than one all the time as they are the smallest modular APIs built in front of end systems.

>> Process APIs can call System APIs as well as other Process APIs. There is no such anti-design pattern in API-Led connectivity saying Process APIs should not call other Process APIs.

So, the right answer in the given set of options that makes sense as per API-Led connectivity principles is to allow System APIs to return data that is NOT currently required by the identified Process or Experience APIs. This way, some future Process APIs can make use of that data from System APIs and we need NOT touch the System layer APIs again and again.

Explanation

Correct Answer: SLA-based rate limiting

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

>> Client Id enforement policy is a "Compliance" related NFR and does not help in maintaining the "Quality of Service (QoS)". It CANNOT and NOT meant for protecting the backend systems from scalability challenges.

>> IP Whitelisting and OAuth 2.0 token enforcement are "Security" related NFRs and again does not help in maintaining the "Quality of Service (QoS)". They CANNOT and are NOT meant for protecting the backend systems from scalability challenges.

Rate Limiting, Rate Limiting-SLA, Throttling, Spike Control are the policies that are "Quality of Service (QOS)" related NFRs and are meant to help in protecting the backend systems from getting overloaded.

https://dzone.com/articles/how-to-secure-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.

In the API-led connectivity approach, each layer (System, Process, and Experience) has a distinct purpose:

System APIs: These APIs connect directly to backend systems to expose and unlock data in a standardized way.

Process APIs: These are responsible for orchestrating and processing data across different systems, combining information where needed.

Experience APIs: These are designed for specific user interfaces or applications, often transforming data formats to fit the needs of each consumer application.

Why Option A is Correct:

Process APIs are designed to combine data from multiple systems, which aligns with the function of aggregating policy information from both the life and general insurance systems. This aggregation logic would ideally reside in the Process layer, separating data retrieval from data orchestration.

Moving this functionality to the Process layer enables reusability and modularity, as other Experience APIs or services could also leverage the combined policy data if needed.

Explanation of Incorrect Options:

Option B (Presenting the SOAP-based response) would be managed by the Experience layer, as this layer adapts data formats for specific interfaces.

Option C (Authenticating and maintaining backend connections) would typically be handled within the System layer, where backend integration and security handling occurs.

Option D (Querying data) is the function of System APIs, which access the backend systems directly and expose the raw data without additional processing.

ReferencesFor further details on API-led architecture and the roles of each layer, refer to MuleSoft’s documentation on API-led connectivity and API layers​​.

Explanation

Correct Answer: The CloudHub shared load balancer

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

The scenario in this question can be split as below:

>> There are 3 CloudHub workers (So, there are already good number of workers to handle high volume of requests)

>> The workers are not using static IP addresses (So, one CANNOT use customer load-balancing solutions without static IPs)

>> Looking for most cost-effective component to load balance the client requests among the workers.

Based on the above details given in the scenario:

>> Runtime autoscaling is NOT at all cost-effective as it incurs extra cost. Most over, there are already 3 workers running which is a good number.

>> We cannot go for a customer-hosted load balancer as it is also NOT most cost-effective (needs custom load balancer to maintain and licensing) and same time the Mule App is not having Static IP Addresses which limits from going with custom load balancing.

>> An API Proxy is irrelevant there as it has no role to play w.r.t handling high volumes or load balancing.

So, the only right option to go with and fits the purpose of scenario being most cost-effective is - using a CloudHub Shared Load Balancer.

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​.