Modeling Data Movement Performance on Heterogeneous Architectures
Modeling Data Movement Performance on Heterogeneous Architectures
复制标题
对异构架构上的数据移动性能进行建模
DOI：
10.1109/hpec49654.2021.9622742
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
S. Lockhart
中科院分区：
文献类型：
--
作者：
Amanda Bienz;Luke N. Olson;W. Gropp;S. Lockhart
并行系统上数据移动的成本因机器架构、任务划分以及相邻任务而有很大差异。能够准确捕捉数据移动成本的性能模型为分析提供了一种工具，从而可以精准定位通信瓶颈。现代异构架构导致数据移动的差异增大，因为GPU间通信存在许多可行的路径。在本文中，我们提出了现代异构架构上节点间各种通信路径的性能模型，包括GPUDirect通信和复制到CPU之间的权衡。此外，我们基于这些模型提出了一种针对节点间通信的新颖优化方法，利用了每个节点上所有可用的CPU核心。最后，我们展示了MPI集合操作相关的性能提升。
The cost of data movement on parallel systems varies greatly with machine architecture, job partition, and nearby jobs. Performance models that accurately capture the cost of data movement provide a tool for analysis, allowing for communication bottlenecks to be pinpointed. Modern heterogeneous architectures yield increased variance in data movement as there are a number of viable paths for inter-GPU communication. In this paper, we present performance models for the various paths of inter-node communication on modern heterogeneous architectures, including the trade-off between GPUDirect communication and copying to CPUs. Furthermore, we present a novel optimization for inter-node communication based on these models, utilizing all available CPU cores per node. Finally, we show associated performance improvements for MPI collective operations.