Two-phase I/O is a well-known strategy for implementing collective MPI-IO functions. It redistributes I/O requests among the calling processes into a form that minimizes the file access costs. As modern parallel computers continue to grow into the exascale era, the communication cost of such request redistribution can quickly overwhelm collective I/O performance. This effect has been observed from parallel jobs that run on multiple compute nodes with a high count of MPI processes on each node. To reduce the communication cost, we present a new design for collective I/O by adding an extra communication layer that performs request aggregation among processes within the same compute nodes. This approach can significantly reduce inter-node communication congestion when redistributing the I/O requests. We evaluate the performance and compare with the original two-phase I/O on a Cray XC40 parallel computer with Intel KNL processors. Using I/O patterns from two large-scale production applications and an I/O benchmark, we show the performance improvement of up to 29 times when running 16384 MPI processes on 256 compute nodes.

中文翻译：

---

使用节点内聚合改进大容量非连续请求的 MPI 集体 I/O

两阶段 I/O 是一种众所周知的用于实现集体 MPI-IO 功能的策略。它将 I/O 请求在调用进程之间重新分配为最小化文件访问成本的形式。随着现代并行计算机继续发展到百亿亿次时代，这种请求重新分配的通信成本很快就会压倒集体 I/O 性能。从在多个计算节点上运行的并行作业中观察到这种影响，每个节点上都有大量 MPI 进程。为了降低通信成本，我们通过添加一个额外的通信层来呈现集体 I/O 的新设计，该层在相同计算节点内的进程之间执行请求聚合。这种方法可以在重新分配 I/O 请求时显着减少节点间通信拥塞。我们在配备英特尔 KNL 处理器的 Cray XC40 并行计算机上评估性能并与原始两相 I/O 进行比较。使用来自两个大规模生产应用程序的 I/O 模式和 I/O 基准测试，我们展示了在 256 个计算节点上运行 16384 个 MPI 进程时性能提升高达 29 倍。