Network optimizations on prediction server with multiple predictors

Kaho Okuyama, Yuta Tokusashi, Takuma Iwata, Mineto Tsukada, Kazumasa Kishiki, Hiroki Matsutani

研究成果: Conference contribution

抄録

Toward machine learning based prediction services, the prediction server has multiple predictors and selects an appropriate one based on past feedbacks from the clients. In this case, three messages including request, reply, and feedback, are required for each prediction request. Packets are typically transmitted and received via a network protocol stack in OS kernel, and performance improvement can be expected by avoiding the protocol stack since it degrades the communication performance especially for small packets. We implement the prediction server using network optimization approaches including kernel-bypassing and in-NIC processing approaches. Evaluation results show that these network optimizations are beneficial to improve the prediction server performance compared to a baseline prediction server using a standard network protocol stack.

本文言語English
ホスト出版物のタイトルProceedings - 16th IEEE International Symposium on Parallel and Distributed Processing with Applications, 17th IEEE International Conference on Ubiquitous Computing and Communications, 8th IEEE International Conference on Big Data and Cloud Computing, 11th IEEE International Conference on Social Computing and Networking and 8th IEEE International Conference on Sustainable Computing and Communications, ISPA/IUCC/BDCloud/SocialCom/SustainCom 2018
編集者Jinjun Chen, Laurence T. Yang
出版社Institute of Electrical and Electronics Engineers Inc.
ページ1044-1045
ページ数2
ISBN(電子版)9781728111414
DOI
出版ステータスPublished - 2019 3 20
イベント16th IEEE International Symposium on Parallel and Distributed Processing with Applications, 17th IEEE International Conference on Ubiquitous Computing and Communications, 8th IEEE International Conference on Big Data and Cloud Computing, 11th IEEE International Conference on Social Computing and Networking and 8th IEEE International Conference on Sustainable Computing and Communications, ISPA/IUCC/BDCloud/SocialCom/SustainCom 2018 - Melbourne, Australia
継続期間: 2018 12 112018 12 13

出版物シリーズ

名前Proceedings - 16th IEEE International Symposium on Parallel and Distributed Processing with Applications, 17th IEEE International Conference on Ubiquitous Computing and Communications, 8th IEEE International Conference on Big Data and Cloud Computing, 11th IEEE International Conference on Social Computing and Networking and 8th IEEE International Conference on Sustainable Computing and Communications, ISPA/IUCC/BDCloud/SocialCom/SustainCom 2018

Conference

Conference16th IEEE International Symposium on Parallel and Distributed Processing with Applications, 17th IEEE International Conference on Ubiquitous Computing and Communications, 8th IEEE International Conference on Big Data and Cloud Computing, 11th IEEE International Conference on Social Computing and Networking and 8th IEEE International Conference on Sustainable Computing and Communications, ISPA/IUCC/BDCloud/SocialCom/SustainCom 2018
国/地域Australia
CityMelbourne
Period18/12/1118/12/13

ASJC Scopus subject areas

  • 計算理論と計算数学
  • コンピュータ サイエンスの応用

フィンガープリント

「Network optimizations on prediction server with multiple predictors」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