Enabling efficient, high-performance data access in parallel file systems (PFS) is critical for today's high-performance computing systems. PFS client-side I/O heavily impacts the final I/O performance delivered to individual applications and the entire system. Autotuning the key client-side I/O behaviors has been extensively studied and shows promising results. However, existing work has heavily relied on extensive number of global runtime metrics to monitor and accurate modeling of applications' I/O patterns. Such heavy overheads significantly limit the ability to enable fine-grained, dynamic tuning in practical systems. In this study, we propose DIAL (Decentralized I/O AutoTuning via Learned Client-side Local Metrics) which takes a drastically different approach. Instead of trying to extract the global I/O patterns of applications, DIAL takes a decentralized approach, treating each I/O client as an independent unit and tuning configurations using only its locally observable metrics. With the help of machine learning models, DIAL enables multiple tunable units to make independent but collective decisions, reacting to what is happening in the global storage systems in a timely manner and achieving better I/O performance globally for the application.

翻译：在并行文件系统中实现高效、高性能的数据访问对当今高性能计算系统至关重要。PFS客户端I/O行为显著影响最终交付给单个应用及整个系统的I/O性能。针对关键客户端I/O行为的自动调优已被广泛研究并展现出良好效果。然而，现有工作严重依赖大量全局运行时指标来监测应用I/O模式并进行精确建模，这种高开销极大限制了在实际系统中实现细粒度动态调优的能力。本研究提出DIAL（基于客户端本地学习指标的并行文件系统去中心化I/O自动调优），采用截然不同的技术路径：不再试图提取应用的全局I/O模式，而是采用去中心化方法，将每个I/O客户端视为独立单元，仅利用其本地可观测指标进行配置调优。借助机器学习模型，DIAL使多个可调优单元能够做出独立而协同的决策，及时响应全局存储系统的动态变化，从而为应用实现更优的全局I/O性能。