Invention:
This invention is in the area of the application of large language models (LLMs), like ChatGPT, to improve the comprehension of microservice architecture within the software domain. Microservices offer flexibility and scalability but can be complex to understand, particularly when dealing with Enterprise Architecture documentation from different teams. This invention improves ChatGPT's ability to answer intricate queries related to cloud-native systems by using both source code and an intermediate representation obtained from static code analysis. Specifically, it addresses questions related to the service and interaction perspectives of microservice systems. Evaluated results demonstrate the benefits of integrating intermediate representations for a richer contextual understanding while acknowledging the model's inherent limitations. Results pave the way for future extensions toward language models, enhancing the comprehension of software in dynamic, microservice-intensive environments.
Background:
Microservices have gained popularity due to their inherent advantages in flexibility and scalability, enabling efficient deployment of software systems. However, this architectural shift has brought about a significant challenge: understanding complex software systems, particularly when dealing with documentation from various teams in Enterprise Architecture. Currently, developers have access to automated code summaries and visual displays, but these tools still necessitate substantial effort to fully grasp the intricacies of microservice systems. This study's innovative approach aims to provide a richer contextual understanding of microservice systems, potentially simplifying the comprehension of dynamic, microservice-heavy settings, differentiating itself from current technologies that often fall short in addressing these complexities.
Applications:
- Software development and engineering
- Cloud-native and microservice-based systems
- Microservice architecture
- Code analysis
Advantages:
- Efficient problem solving
- Enhanced comprehension
- Richer contextual understanding
