Migrating MuleSoft API led connectivity pattern into AWS using Kumologica
API led connectivity pattern is the popular design pattern followed by MuleSoft integration developers. It’s a methodology pertains to the connection and integration of various applications and systems through APIs (Application Programming Interfaces).
In this article we will be seeing how API led design pattern is visualized in AWS using Kumologica. Kumologica facilitates rapid development and deployment of serverless services through a low-code approach. For those who are new to Kumologica I would recommend to go through our articles and youtube videos to get an insight.
Training & Certification
For a limited time, Kumologica offers free online certification
for developers on https://training.kumologica.com/Layers of API led integration pattern
API led pattern involves 3 layers. An experience layer, process layer and a system layer. Lets go through briefly on these layers.
Systems API
These represent the most basic APIs, facilitating connections with individual systems or applications. They offer a standardized interface for accessing the functionalities of the underlying systems, simplifying their integration into the broader architecture.
Process API
These APIs coordinate various system APIs to generate fresh business processes or workflows. Their role involves delineating the sequence of actions needed to accomplish a task or transaction.
Experience API
These represent the top-tier APIs offering a cohesive interface to external users, like mobile apps or web applications. They consolidate data and services from various underlying systems, ensuring a uniform user experience across diverse channels.
Use case
An enterprise is having a web application and a mobile application requesting data from the same set of backend systems eg . Salesforce, Database. The response to web and mobile applications are aggregated data form Salesforce and database.
Architecture
In MuleSoft and Anypoint platform
The use case is implemented in MuleSoft with Anypoint platform and Mule Runtime. The design layout is given below.
In the above design, the experience layer is having mobile and web experience APIs are two services running as independent MuleSoft workers . The workers are exposed to the users via MuleSoft API gateway. The security the API is applied from the Anypoint API gateway. Similarly the process layer is having client process API which is carrying the business logic for aggregating the data from salesforce and database. Finally we have the system layer which is having salesforce system API and database API which are interacting with the Salesforce and database server respectively. MuleSoft flow uses the Salesforce and data base connector in in system API to connect with the systems. These API executes the CRUD action based on the logic from process layer.
In Kumologica and AWS
In the above design architecture, the 3 layers are maintained exactly same as in API led architecture. Mobile experience and web experience APIs are developed on Kumologica and is deployed on AWS lambda. The lambdas are exposed to the end users via AWS API Gateway with necessary security authorizers. The experience layer will be communicating to process layer via lambda sync invoke. This actually avoids unnecessary hopping over the API gateway for intra-service communication. The client process will have its orchestration defined in Kumologica flow exactly similar to Mulesoft flow. The client process API invokes system layer API via lambda sync invoke. The salesforce system API and database system API will have the Kumologica flow having the salesforce node and database node which will be interacting with the appropriate systems.
Technology alternatives
Following are the technology alternatives considered when undergoing migration to AWS serverless world.
Conclusion
This article has covered the approach from the design perspective. For deep dive on the implementation you may check out our earlier article on MuleSoft migrations.
Remember Kumologica is totally free to download and use. Go ahead and give it a try, we would love to hear your feedback
