A memory congestion-aware MPI Process Placement for Modern NUMA Systems

Mulya Agung, Muhammad Alfian Amrizal, Kazuhiko Komatsu, Ryusuke Egawa, Hiroyuki Takizawa

研究成果: Conference contribution

5 被引用数 (Scopus)

抄録

MPI process placement is an important step to achieve scalable performance on modern non-uniform memory access (NUMA) systems. A recent study on NUMA architectures has shown that, on modern NUMA systems, the memory congestion problem could cause more severe performance degradation than the data locality problem because heavy congestion on memory controllers could cause long latencies. However, conventional work on MPI process placement has focused on locality to minimize the remote-access communication. Moreover, maximizing the locality may actually degrade performance because the load imbalance among nodes in a modern NUMA system may increase. Thus, a process placement algorithm must be designed to consider memory congestion. In this paper, a method to reconcile both the locality and the memory congestion on modern NUMA systems is proposed. This method statically analyzes the application communication pattern to optimize the process placement. A data clustering method is applied to the time-series data of the MPI communications in order to identify data traffics that potentially cause memory congestion. The proposed method has been evaluated with the NPB kernels on a real NUMA system and a simulation environment. Experimental results show that the proposed method can achieve 1.6x performance improvement compared with the current state-of-the-art strategy.

本文言語English
ホスト出版物のタイトルProceedings - 24th IEEE International Conference on High Performance Computing, HiPC 2017
出版社Institute of Electrical and Electronics Engineers Inc.
ページ152-161
ページ数10
ISBN(電子版)9781538622933
DOI
出版ステータスPublished - 2018 2 7
イベント24th IEEE International Conference on High Performance Computing, HiPC 2017 - Jaipur, India
継続期間: 2017 12 182017 12 21

出版物シリーズ

名前Proceedings - 24th IEEE International Conference on High Performance Computing, HiPC 2017
2017-December

Other

Other24th IEEE International Conference on High Performance Computing, HiPC 2017
国/地域India
CityJaipur
Period17/12/1817/12/21

ASJC Scopus subject areas

  • コンピュータ ネットワークおよび通信
  • ハードウェアとアーキテクチャ
  • モデリングとシミュレーション

フィンガープリント

「A memory congestion-aware MPI Process Placement for Modern NUMA Systems」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