DarwinGame: Playing Tournaments for Tuning Applications in Noisy Cloud Environments

2025-09-29

AI Summary

In shared cloud environments, resource interference severely degrades the robustness and accuracy of existing performance tuning tools. To address this, we propose a tournament-based application performance tuning method—the first to introduce relative performance comparison into automated tuning under noisy conditions. Our approach employs iterative pairwise configuration comparisons, robust execution scheduling, and statistical stability enhancement to effectively suppress interference effects. Compared to state-of-the-art tuners, it reduces execution time by over 27% and constrains performance variability to within 0.5%. The method achieves superior efficiency, precision, and robustness simultaneously, establishing a novel paradigm for automated performance optimization in interference-prone cloud environments.

Technology Category

• Search and Optimization: Algorithm Configuration
• Planning, Routing, and Scheduling: Applications
• Machine Learning: Auto ML and Hyperparameter Tuning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

This work introduces a new subarea of performance tuning -- performance tuning in a shared interference-prone computing environment. We demonstrate that existing tuners are significantly suboptimal by design because of their inability to account for interference during tuning. Our solution, DarwinGame, employs a tournament-based design to systematically compare application executions with different tunable parameter configurations, enabling it to identify the relative performance of different tunable parameter configurations in a noisy environment. Compared to existing solutions, DarwinGame achieves more than 27% reduction in execution time, with less than 0.5% performance variability. DarwinGame is the first performance tuner that will help developers tune their applications in shared, interference-prone, cloud environments.

Problem

• Tuning applications in shared noisy cloud environments
• Addressing performance interference during parameter optimization
• Systematically comparing configurations in interference-prone computing environments

Innovation

• Tournament-based design for parameter comparison
• Systematic performance evaluation in noisy environments
• Reduces execution time with minimal performance variability