HPC Application Parameter Autotuning on Edge Devices: A Bandit Learning Approach

2025-01-02

AI Summary

Automated parameter tuning of high-performance computing (HPC) applications on resource-constrained edge devices—such as smartphones and embedded terminals—remains challenging due to stringent latency and memory constraints. Method: This paper proposes LASP, a lightweight online adaptive tuning framework that introduces a novel dynamic multi-armed bandit (MAB) paradigm tailored for edge environments. Unlike conventional offline or high-overhead tuning approaches, LASP integrates online reinforcement learning with lightweight runtime search, enabling efficient parameter exploration and environment adaptation under sub-millisecond response times and KB-scale memory budgets. Contribution/Results: Evaluated on four representative HPC benchmarks—Lulesh, Kripke, Clomp, and Hypre—LASP achieves significant performance improvements over baseline methods while reducing computational and memory overhead by over 90%. It thus overcomes the critical bottlenecks of low-latency, low-resource tuning at the edge.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

The growing necessity for enhanced processing capabilities in edge devices with limited resources has led us to develop effective methods for improving high-performance computing (HPC) applications. In this paper, we introduce LASP (Lightweight Autotuning of Scientific Application Parameters), a novel strategy designed to address the parameter search space challenge in edge devices. Our strategy employs a multi-armed bandit (MAB) technique focused on online exploration and exploitation. Notably, LASP takes a dynamic approach, adapting seamlessly to changing environments. We tested LASP with four HPC applications: Lulesh, Kripke, Clomp, and Hypre. Its lightweight nature makes it particularly well-suited for resource-constrained edge devices. By employing the MAB framework to efficiently navigate the search space, we achieved significant performance improvements while adhering to the stringent computational limits of edge devices. Our experimental results demonstrate the effectiveness of LASP in optimizing parameter search on edge devices.

Problem

• Automatic Optimization
• Complex Applications
• Resource-limited Devices

Innovation

• LASP
• Multi-Armed Bandit
• Resource-Constrained Devices