A Combined Autotuning and Runtime Resource Management Framework for Dynamic Workload Optimization on Homogeneous Architectures
Modern computing systems operate under highly variable and unpredictable workloads, with performance-critical applications constantly entering and leaving the execution environment. These fluctuations create challenges in meeting strict performance guarantees while maintaining energy efficiency. Traditional approaches typically rely on either application-level autotuning or architecture-level resource management, but these independent strategies often fall short in addressing complex performance–power–quality trade-offs. This study introduces an integrated two-level framework that unifies both mechanisms to improve system responsiveness, efficiency, and reliability on homogeneous architectures. Background and Motivation As computing platforms grow increasingly heterogeneous in their application demands—even on homogeneous hardware—they face the dual challenge of sustaining application performance and minimizing power consumption. Autotuners can optimize software-level parameters suc...