Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks

Jianping Kelvin Li, Misbah Mubarak, Robert B. Ross, Christopher D. Carothers, Kwan-Liu Ma

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

High-radix, low-diameter, hierarchical networks based on the Dragonfly topology are common picks for building next generation HPC systems. However, effective tools are lacking for analyzing the network performance and exploring the design choices for such emerging networks at scale. In this paper, we present visual analytics methods that couple data aggregation techniques with interactive visualizations for analyzing large-scale Dragonfly networks. We create an interactive visual analytics system based on these techniques. To facilitate effective analysis and exploration of network behaviors, our system provides intuitive, scalable visualizations that can be customized to show various traffic characteristics and correlate between different performance metrics. Using high-fidelity network simulation and HPC applications communication traces, we demonstrate the usefulness of our system with several case studies on exploring network behaviors at scale with different workloads, routing strategies, and job placement policies. Our simulations and visualizations provide valuable insights for mitigating network congestion and inter-job interference.

Original languageEnglish (US)
Title of host publicationProceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages193-203
Number of pages11
Volume2017-September
ISBN (Electronic)9781538623268
DOIs
StatePublished - Sep 22 2017
Event2017 IEEE International Conference on Cluster Computing, CLUSTER 2017 - Honolulu, United States
Duration: Sep 5 2017Sep 8 2017

Other

Other2017 IEEE International Conference on Cluster Computing, CLUSTER 2017
CountryUnited States
CityHonolulu
Period9/5/179/8/17

Fingerprint

Visualization
Network performance
Agglomeration
Topology
Communication

Keywords

  • Dragonfly networks
  • Performance analysis
  • Visual analytics
  • Visualization

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Signal Processing

Cite this

Li, J. K., Mubarak, M., Ross, R. B., Carothers, C. D., & Ma, K-L. (2017). Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks. In Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017 (Vol. 2017-September, pp. 193-203). [8048931] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CLUSTER.2017.26

Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks. / Li, Jianping Kelvin; Mubarak, Misbah; Ross, Robert B.; Carothers, Christopher D.; Ma, Kwan-Liu.

Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017. Vol. 2017-September Institute of Electrical and Electronics Engineers Inc., 2017. p. 193-203 8048931.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Li, JK, Mubarak, M, Ross, RB, Carothers, CD & Ma, K-L 2017, Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks. in Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017. vol. 2017-September, 8048931, Institute of Electrical and Electronics Engineers Inc., pp. 193-203, 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017, Honolulu, United States, 9/5/17. https://doi.org/10.1109/CLUSTER.2017.26
Li JK, Mubarak M, Ross RB, Carothers CD, Ma K-L. Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks. In Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017. Vol. 2017-September. Institute of Electrical and Electronics Engineers Inc. 2017. p. 193-203. 8048931 https://doi.org/10.1109/CLUSTER.2017.26
Li, Jianping Kelvin ; Mubarak, Misbah ; Ross, Robert B. ; Carothers, Christopher D. ; Ma, Kwan-Liu. / Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks. Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017. Vol. 2017-September Institute of Electrical and Electronics Engineers Inc., 2017. pp. 193-203
@inproceedings{876a22b7fcbb427c9117fbb336e39345,
title = "Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks",
abstract = "High-radix, low-diameter, hierarchical networks based on the Dragonfly topology are common picks for building next generation HPC systems. However, effective tools are lacking for analyzing the network performance and exploring the design choices for such emerging networks at scale. In this paper, we present visual analytics methods that couple data aggregation techniques with interactive visualizations for analyzing large-scale Dragonfly networks. We create an interactive visual analytics system based on these techniques. To facilitate effective analysis and exploration of network behaviors, our system provides intuitive, scalable visualizations that can be customized to show various traffic characteristics and correlate between different performance metrics. Using high-fidelity network simulation and HPC applications communication traces, we demonstrate the usefulness of our system with several case studies on exploring network behaviors at scale with different workloads, routing strategies, and job placement policies. Our simulations and visualizations provide valuable insights for mitigating network congestion and inter-job interference.",
keywords = "Dragonfly networks, Performance analysis, Visual analytics, Visualization",
author = "Li, {Jianping Kelvin} and Misbah Mubarak and Ross, {Robert B.} and Carothers, {Christopher D.} and Kwan-Liu Ma",
year = "2017",
month = "9",
day = "22",
doi = "10.1109/CLUSTER.2017.26",
language = "English (US)",
volume = "2017-September",
pages = "193--203",
booktitle = "Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Visual Analytics Techniques for Exploring the Design Space of Large-Scale High-Radix Networks

AU - Li, Jianping Kelvin

AU - Mubarak, Misbah

AU - Ross, Robert B.

AU - Carothers, Christopher D.

AU - Ma, Kwan-Liu

PY - 2017/9/22

Y1 - 2017/9/22

N2 - High-radix, low-diameter, hierarchical networks based on the Dragonfly topology are common picks for building next generation HPC systems. However, effective tools are lacking for analyzing the network performance and exploring the design choices for such emerging networks at scale. In this paper, we present visual analytics methods that couple data aggregation techniques with interactive visualizations for analyzing large-scale Dragonfly networks. We create an interactive visual analytics system based on these techniques. To facilitate effective analysis and exploration of network behaviors, our system provides intuitive, scalable visualizations that can be customized to show various traffic characteristics and correlate between different performance metrics. Using high-fidelity network simulation and HPC applications communication traces, we demonstrate the usefulness of our system with several case studies on exploring network behaviors at scale with different workloads, routing strategies, and job placement policies. Our simulations and visualizations provide valuable insights for mitigating network congestion and inter-job interference.

AB - High-radix, low-diameter, hierarchical networks based on the Dragonfly topology are common picks for building next generation HPC systems. However, effective tools are lacking for analyzing the network performance and exploring the design choices for such emerging networks at scale. In this paper, we present visual analytics methods that couple data aggregation techniques with interactive visualizations for analyzing large-scale Dragonfly networks. We create an interactive visual analytics system based on these techniques. To facilitate effective analysis and exploration of network behaviors, our system provides intuitive, scalable visualizations that can be customized to show various traffic characteristics and correlate between different performance metrics. Using high-fidelity network simulation and HPC applications communication traces, we demonstrate the usefulness of our system with several case studies on exploring network behaviors at scale with different workloads, routing strategies, and job placement policies. Our simulations and visualizations provide valuable insights for mitigating network congestion and inter-job interference.

KW - Dragonfly networks

KW - Performance analysis

KW - Visual analytics

KW - Visualization

UR - http://www.scopus.com/inward/record.url?scp=85032644966&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032644966&partnerID=8YFLogxK

U2 - 10.1109/CLUSTER.2017.26

DO - 10.1109/CLUSTER.2017.26

M3 - Conference contribution

AN - SCOPUS:85032644966

VL - 2017-September

SP - 193

EP - 203

BT - Proceedings - 2017 IEEE International Conference on Cluster Computing, CLUSTER 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -