INTEGRATION-ARCHITECTURE-DESIGNER practice question 19 of 25

Universal Containers (UC) is currently managing a custom monolithic web service that runs on an on- premise server. This monolithic web service is…

Full set: 106 questions

Domain Design an Integration Solution

Universal Containers (UC) is currently managing a custom monolithic web service that runs on an on- premise server. This monolithic web service is responsible for Point-to-Point (P2P) integrations between: 1. Salesforce and a legacy billing application 2. Salesforce and a cloud-based Enterprise Resource Planning application 3. Salesforce and a data lake. UC has found that the tight interdependencies between systems is causing integrations to fail. What should an architect recommend to decouple the systems and improve performance of the integrations?
Answer options
A: Re-write and optimize the current web service to be more efficient.
B: Leverage modular design by breaking up the web service into smaller pieces for a microservice architecture. Correct answer
C: Use the Salesforce Bulk API when integrating back into Salesforce.
D: Move the custom monolithic web service from on-premise to a cloud provider.
Correct Answer:
B. Leverage modular design by breaking up the web service into smaller pieces for a microservice architecture.
Explanation

The core problem is the "tight interdependencies" of a monolithic web service causing failures. A microservice architecture directly addresses this by breaking down the single, large service into smaller, independent, and loosely coupled services. Each microservice can manage a specific integration (e.g., one for billing, one for ERP). This modularity decouples the systems, meaning a failure or change in one integration service does not impact the others. This approach enhances resilience, simplifies maintenance, and allows for independent scaling, which improves overall performance and reliability.

Why Incorrect

A. Re-write and optimize the current web service to be more efficient. This might improve performance but fails to address the fundamental architectural flaw of tight coupling. The system would remain a monolith, and interdependencies would persist.
C. Use the Salesforce Bulk API when integrating back into Salesforce. This is a tactical optimization for large data volume scenarios. It does not solve the strategic architectural problem of the monolithic service and its tight interdependencies.
D. Move the custom monolithic web service from on-premise to a cloud provider. This is an infrastructure change ("lift and shift"), not an architectural one. The application would still be a monolith with the same tight coupling issues, regardless of where it is hosted.

References

1. Salesforce Architects - Integration Patterns and Practices Guide: This guide discusses the challenges of point-to-point integrations, which become "brittle" and hard to maintain as complexity grows—a direct symptom of tight coupling. It advocates for patterns that decouple systems. A microservices approach is a modern implementation of this principle. (See "Introduction" and "Remote Process Invocation—Request and Reply" sections for context on point-to-point challenges).

2. Salesforce Architects - Architect's Guide to Integration: This guide emphasizes the importance of choosing the right architecture to avoid the pitfalls of tightly coupled systems. It states, "Point-to-point integrations are brittle and create a complex web of dependencies that is difficult to maintain and change." (See "Integration Architecture Patterns" section). Breaking a monolith into microservices is a direct solution to this described problem.

3. Martin, R. C. (2017). Clean Architecture: A Craftsman's Guide to Software Structure and Design. Prentice Hall. While not a Salesforce-specific document, this foundational software engineering text explains the Single Responsibility Principle. Applying this principle to the monolithic service suggests breaking it into smaller components (microservices), each with a single responsibility, thereby achieving decoupling. (See Chapter 7: "SRP: The Single Responsibility Principle").

Scroll to Top