Evaluation of Topology-Aware Broadcast Algorithms for Dragonfly Networks

Conference · Mon Sep 12 00:00:00 EDT 2016

DOI:https://doi.org/10.1109/CLUSTER.2016.26· OSTI ID:1366302

Dorier, Matthieu; Mubarak, Misbah; Ross, Rob; Li, Jianping Kelvin; Carothers, Christopher D.; Ma, Kwan-Liu

Two-tiered direct network topologies such as Dragonflies have been proposed for future post-petascale and exascale machines, since they provide a high-radix, low-diameter, fast interconnection network. Such topologies call for redesigning MPI collective communication algorithms in order to attain the best performance. Yet as increasingly more applications share a machine, it is not clear how these topology-aware algorithms will react to interference with concurrent jobs accessing the same network. In this paper, we study three topology-aware broadcast algorithms, including one designed by ourselves. We evaluate their performance through event-driven simulation for small- and large-sized broadcasts (in terms of both data size and number of processes). We study the effect of different routing mechanisms on the topology-aware collective algorithms, as well as their sensitivity to network contention with other jobs. Our results show that while topology-aware algorithms dramatically reduce link utilization, their advantage in terms of latency is more limited.

• Research Organization: Argonne National Lab. (ANL), Argonne, IL (United States)
• Sponsoring Organization: USDOE Office of Science - Office of Advanced Scientific Computing Research
• DOE Contract Number: AC02-06CH11357
• OSTI ID: 1366302
• Resource Relation: Conference: 2016 Institute of Electrical and Electronics Engineers Cluster, 09/12/16 - 09/16/16, Taipei, TW
• Country of Publication: United States
• Language: English