Show Cover Slide

Case Study: Revamping the Edge + Cloud Platform Architecture to Secure a Technological Foundation and Enable Scalable Replication

gpt-4-turbo has translated this article into English.

Redesigning an Initially Implemented System into a Scalable Architecture and Verifying the Feasibility for Platform Reusability

Problem Recognition

Various edge devices such as access control units and temperature controllers were operating independently, with an imperfect connection structure to the cloud. Although functional, there was a lack of scalability, maintainability, and consistency in the developer experience.

My Design Philosophy

I believe that the essence of a platform is not “device connection” but “consistently organizing the flow and operational experience between devices.” I prioritize design focused on structure and flow, rather than technology and features.

Design Direction

In this project, the goal was not just “implementing functions” but designing it to be a platform.

• Structured MQTT-based messaging pipeline
• Introduction of AWS IoT Rule, Device Shadow, Greengrass → Separation of state management and event handling
• Docker-based edge operating environment → Ensuring a consistent deployment environment
• Schema-based message structure definition → Ensuring compatibility and scalability between devices
• Elimination and refactoring of existing technical debt (hardcoding, inconsistent logic)

Results

• Organized the flow between cloud and edge,
• Structured communication among various devices in a uniform manner,
• Transitioned to a reusable structure for new device introductions.
• Most importantly, secured a platform basis that can be repetitively applied to multiple spaces with a single design.

In Conclusion

What I wanted to emphasize through this project was not just ‘cleaning up code’ but ‘creating a structure that can expand business’. While creating features is important, designing a foundation that can be expanded is what I believe to be the role of an architect.