Quantifying the impact of network congestion on application performance and network metrics
Authors:
- 1 September 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 162-168
Abstract
In modern high-performance computing (HPC) systems, network congestion is an important factor that contributes to performance degradation. However, how network congestion impacts application performance is not fully understood. As Aries network, a recent HPC network architecture featuring a dragonfly topology, is equipped with network counters measuring packet transmission statistics on each router, these network metrics can potentially be utilized to understand network performance. In this work, by experiments on a large HPC system, we quantify the impact of network congestion on various applications' performance in terms of execution time, and we correlate application performance with network metrics. Our results demonstrate diverse impacts of network congestion: while applications with intensive MPI operations (such as HACC and MILC) suffer from more than 40% extension in their execution times under network congestion, applications with less intensive MPI operations (such as Graph500 and HPCG) are mostly not affected. We also demonstrate that a stall-to-flit ratio metric derived from Aries network counters is positively correlated with performance degradation and, thus, this metric can serve as an indicator of network congestion in HPC systems.Keywords
This publication has 24 references indexed in Scilit:
- Understanding Performance Variability on the Aries Dragonfly Network2017
- New Link Arrangements for Dragonfly Networks2017
- Traffic Pattern-Based Adaptive Routing for Intra-Group Communication in Dragonfly Networks2016
- Watch Out for the Bully! Job Interference Study on Dragonfly Network2016
- Comparing Global Link Arrangements for Dragonfly Networks2015
- Identifying the Culprits Behind Network Congestion2015
- The Lightweight Distributed Metric Service: A Scalable Infrastructure for Continuous Monitoring of Large Scale Computing Systems and Applications2014
- There goes the neighborhood2013
- Performance Modeling and Comparative Analysis of the MILC Lattice QCD Application su3_rmd2012
- Indirect adaptive routing on large scale interconnection networksACM SIGARCH Computer Architecture News, 2009