The AI hardware boom has led modern data centers to adopt HPC-style architectures centered on distributed, GPU-centric computation. Large GPU clusters interconnected by fast RDMA networks and backed by high-bandwidth NVMe storage enable scalable computation and rapid access to storage-resident data. Tensor computation runtimes (TCRs), such as PyTorch, originally designed for AI workloads, have recently been shown to accelerate analytical workloads. However, prior work has primarily considered settings where the data fits in aggregated GPU memory. In this paper, we systematically study how TCRs can support scalable, distributed query processing for large-scale, storage-resident OLAP workloads. Although TCRs provide abstractions for network and storage I/O, naive use often underutilizes GPU and I/O bandwidth due to insufficient overlap between computation and data movement. As a core contribution, we present PystachIO, a prototype of a PyTorch-based distributed OLAP engine that combines fast network and storage I/O with key optimizations to maximize GPU, network, and storage utilization. Our evaluation shows up to 3x end-to-end speedups over existing distributed GPU-based query processing approaches.

翻译：AI硬件热潮促使现代数据中心采用以分布式、GPU中心化计算为核心的高性能计算架构。由快速RDMA网络互联并配备高带宽NVMe存储的大型GPU集群，实现了可扩展计算与存储驻留数据的快速访问。原本为AI工作负载设计的张量计算运行时（TCR，如PyTorch）近期已被证明可加速分析型工作负载。然而，先前研究主要考虑数据适合聚合GPU内存的场景。本文系统研究了TCR如何支持大规模、存储驻留OLAP工作负载的可扩展分布式查询处理。尽管TCR提供了网络和存储I/O的抽象，但由于计算与数据移动之间的重叠不足，直接使用往往会导致GPU和I/O带宽利用率低下。作为核心贡献，我们提出了PystachIO——一个基于PyTorch的分布式OLAP引擎原型，它结合了高速网络与存储I/O，并通过关键优化最大化GPU、网络和存储的利用率。我们的评估显示，与现有基于分布式GPU的查询处理方法相比，其端到端速度提升可达3倍。