Gossip: Efficient Communication Primitives for Multi-GPU Systems
Gossip: Efficient Communication Primitives for Multi-GPU Systems
复制标题
DOI：
10.1145/3337821.3337889
复制
发表时间：
2019-08
期刊：
影响因子：
--
通讯作者：
Robin Kobus;Daniel Jünger;Christian Hundt;B. Schmidt
中科院分区：
文献类型：
--
作者：
Robin Kobus;Daniel Jünger;Christian Hundt;B. Schmidt
如今，越来越多的服务器和工作站配备了越来越多的GPU。但是，GPU之间的通信速度较慢可能会导致应用程序性能较差。因此，在这样的系统上存在对高效的多GPU通信原语的潜在需求。本文重点讨论了聚集、分散和All-to-All集合，它们是并行排序和分布式哈希等各种算法的重要操作。我们提出了两种不同的通信策略(基于环的和面向流的)来为它们在NVLink连接的多GPU系统上的拓扑感知实现生成传输计划。我们在一台DGX-1服务器上以很小的内存开销实现了高达526 GB/S的吞吐量和148 GB/S的分散/聚集吞吐量。此外，我们提出了一种DGX-1Volta拓扑的成本中立的替代方案，该拓扑为All-to-All集合提供了预期的更高吞吐量，同时在分散/聚集的情况下保持了吞吐量。我们的八卦图书馆可以在https://github.com/Funatiq/gossip.上免费获得
Nowadays, a growing number of servers and workstations feature an increasing number of GPUs. However, slow communication among GPUs can lead to poor application performance. Thus, there is a latent demand for efficient multi-GPU communication primitives on such systems. This paper focuses on the gather, scatter and all-to-all collectives, which are important operations for various algorithms including parallel sorting and distributed hashing. We present two distinct communication strategies (ring-based and flow-oriented) to generate transfer plans for their topology-aware implementation on NVLink-connected multi-GPU systems. We achieve a throughput of up to 526 GB/s for all-to-all and 148 GB/s for scatter/gather on a DGX-1 server with only a small memory overhead. Furthermore, we propose a cost-neutral alternative to the DGX-1 Volta topology that provides an expected higher throughput for the all-to-all collective while preserving the throughput in case of scatter/gather. Our Gossip library is freely available at https://github.com/Funatiq/gossip.