What is it about?
Software configuration is critical for adjusting software performance to meet different user requirements. Deep learning has emerged as an emerging technique for software configuration and performance modeling, yet a comprehensive summary remains lacking.
Featured Image
Photo by Elisa Calvet B. on Unsplash
Why is it important?
This work surveys 99 relevant studies to systematically examine the four key stages in DL-based configuration performance modeling (data preparation, model training, evaluation, and application), highlighting key trends and common challenges in the literature, and offering actionable insights and future directions.
Perspectives
I hope this study serves as a valuable starting point and provides guidelines for new researchers, while also sheds light on future paths and inspires fresh ideas for experts in the field.
Jingzhi Gong
University of Leeds
Read the Original
This page is a summary of: Deep Configuration Performance Learning: A Systematic Survey and Taxonomy, ACM Transactions on Software Engineering and Methodology, November 2024, ACM (Association for Computing Machinery),
DOI: 10.1145/3702986.
You can read the full text:
Contributors
The following have contributed to this page
