Two papers accepted at the International Conference on Software Architecture (ICSA) 2026

Two papers accepted at the International Conference on Software Architecture (ICSA) 2026

Green Autoscaler for Performance Aware Microservices: a Machine Learning Approach
Thanh-Phuc Tran , Abhinandan Roul , Ishara Galbokka Hewage , Mahira Joytu , Roberta Capuano, Eoan O’Dea, Rafiullah Omar, Hergys Rexha, Sebastien Lanfond , Henry Muccini

LLMs for Architectural Refactoring: An Exploratory study on Monoliths to Microservices
Aneesh Sambu, Roberta Capuano, Eoan O’Dea, Karthik Vaidhyanathan, Henry Muccini

We are thrilled to announce that two of our research papers have been accepted at the International Conference on Software Architecture (ICSA 2026)!

Our first work, “Green Autoscaler for Performance Aware Microservices: a Machine Learning Approach,” tackles the environmental impact of cloud-native systems. Traditional tools like the Kubernetes Horizontal Pod Autoscaler often ignore the carbon intensity of electricity, leading to excessive emissions. We propose a Carbon-Aware Autoscaling (CAAS) system using Spatio-Temporal Graph Convolutional Networks (STGCN) to model workload dynamics and inter-service dependencies. By integrating real-time carbon intensity signals, our approach achieves up to a 24% reduction in carbon emissions while maintaining high performance.
Our second paper, “LLMs for Architectural Refactoring: An Exploratory Study on Monoliths to Microservices,” explores how AI can support the complex journey of system migration. Decomposing monoliths is error-prone, so we developed an LLM-driven pipeline to help architects identify meaningful service boundaries. Our study reveals that LLMs are exceptional at producing architecturally clean designs, outperforming traditional tools in Cyclic Independence (+22.7). Rather than replacing human experts, LLMs act as powerful “architecture design partners,” providing technically coherent alternatives for better dependency management.

Join us at ICSA 2026 to discuss how AI and graph learning are shaping a sustainable and modular future for software architecture! 🌍💻

https://conf.researchr.org/home/icsa-2026

Categories: News

Leave a Reply

Your email address will not be published. Required fields are marked *