🤖 AI Summary
Cloud architecture design faces challenges including ambiguous requirements, implicit decision-making processes, and complex interdependencies among architectural elements. To address these, this paper introduces CloudArchitectBuddy—a novel system featuring a synergistic dual-mechanism framework: *structured state management* and *guided decision support*. It explicitly models requirement evolution and design decisions via a finite-state machine, integrates rule-based validation with large language models (LLMs) for requirement refinement and consistency checking, and establishes a hybrid human–AI collaboration paradigm combining structured workflows with natural-language dialogue. Evaluated with 16 industry practitioners, CloudArchitectBuddy achieves design quality comparable to pure chat-based interfaces while significantly improving usability (*p* < 0.01), architectural relationship comprehension (+32%), and missing requirement identification (+41%). These results empirically validate the effectiveness and practical value of its mixed-interaction approach.
📝 Abstract
Cloud architecture design is a complex process requiring both technical expertise and architectural knowledge to develop solutions from frequently ambiguous requirements. We present CloudArchitectBuddy, a system-driven cloud architecture design support application with two key mechanisms: (1) structured state management that enhances design understanding through explicit representation of requirements and architectural decisions, and (2) guided decision assistance that facilitates design progress through proactive verification and requirement refinement. Our study with 16 industry practitioners showed that while our approach achieved comparable design quality to a chat interface, participants rated our system higher for usability and appreciated its ability to help understand architectural relationships and identify missing requirements. However, participants also expressed a need for user-initiated interactions where they could freely provide design instructions and engage in detailed discussions with LLMs. These results suggest that integrating a chat interface into our structured and guided workflow approach would create a more practical solution, balancing systematic design support with conversational flexibility for comprehensive cloud architecture development.