Evaluating the Impact of Energy Efficient Networks on HPC Workloads

Giorgis Georgakoudis, Nikhil Jain, Takatsugu Ono, Koji Inoue, Shinobu Miwa, Abhinav Bhatele

研究成果: Chapter in Book/Report/Conference proceedingConference contribution

4 被引用数 (Scopus)

抄録

Interconnection networks grow larger as supercomputers include more nodes and require higher bandwidth for performance. This scaling significantly increases the fraction of power consumed by the network, by increasing the number of network components (links and switches). Typically, network links consume power continuously once they are turned on. However, recent proposals for energy efficient interconnects have introduced low-power operation modes for periods when network links are idle. Low-power operation can increase messaging time when switching a link from low-power to active operation. We extend the TraceR-CODES network simulator for power modeling to evaluate the impact of energy efficient networking on power and performance. Our evaluation presents the first study on both single-job and multi-job execution to realistically simulate power consumption and performance under congestion for a large-scale HPC network. Results on several workloads consisting of HPC proxy applications show that single-job and multi-job execution favor different modes of low power operation to have significant power savings at the cost of minimal performance degradation.

本文言語英語
ホスト出版物のタイトルProceedings - 26th IEEE International Conference on High Performance Computing, HiPC 2019
出版社Institute of Electrical and Electronics Engineers Inc.
ページ301-310
ページ数10
ISBN(電子版)9781728145358
DOI
出版ステータス出版済み - 12 2019
イベント26th Annual IEEE International Conference on High Performance Computing, HiPC 2019 - Hyderabad, インド
継続期間: 12 17 201912 20 2019

出版物シリーズ

名前Proceedings - 26th IEEE International Conference on High Performance Computing, HiPC 2019

会議

会議26th Annual IEEE International Conference on High Performance Computing, HiPC 2019
国/地域インド
CityHyderabad
Period12/17/1912/20/19

All Science Journal Classification (ASJC) codes

  • ハードウェアとアーキテクチャ
  • 情報システム
  • 情報システムおよび情報管理
  • 安全性、リスク、信頼性、品質管理

フィンガープリント

「Evaluating the Impact of Energy Efficient Networks on HPC Workloads」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル