hal.inria.fr/inria-00486178

Preview meta tags from the hal.inria.fr website.

Linked Hostnames

24

Search Engine Appearance

Google

https://hal.inria.fr/inria-00486178

Adaptive MPI Multirail Tuning for Non-Uniform Input/Output Access

Multicore processors have not only reintroduced Non-Uniform Memory Access (NUMA) architectures in nowadays parallel computers, but they are also responsible for non-uniform access times with respect to Input/Output devices (NUIOA). In clusters of multicore machines equipped with several Network Interfaces, performance of communication between processes thus depends on which cores these processes are scheduled on, and on their distance to the Network Interface Cards involved. We propose a technique allowing multirail communication between processes to carefully distribute data among the network interfaces so as to counterbalance NUIOA effects. We demonstrate the relevance of our approach by evaluating its implementation within OpenMPI on a Myri-10G + InfiniBand cluster.



Bing

Adaptive MPI Multirail Tuning for Non-Uniform Input/Output Access

https://hal.inria.fr/inria-00486178

Multicore processors have not only reintroduced Non-Uniform Memory Access (NUMA) architectures in nowadays parallel computers, but they are also responsible for non-uniform access times with respect to Input/Output devices (NUIOA). In clusters of multicore machines equipped with several Network Interfaces, performance of communication between processes thus depends on which cores these processes are scheduled on, and on their distance to the Network Interface Cards involved. We propose a technique allowing multirail communication between processes to carefully distribute data among the network interfaces so as to counterbalance NUIOA effects. We demonstrate the relevance of our approach by evaluating its implementation within OpenMPI on a Myri-10G + InfiniBand cluster.



DuckDuckGo

https://hal.inria.fr/inria-00486178

Adaptive MPI Multirail Tuning for Non-Uniform Input/Output Access

Multicore processors have not only reintroduced Non-Uniform Memory Access (NUMA) architectures in nowadays parallel computers, but they are also responsible for non-uniform access times with respect to Input/Output devices (NUIOA). In clusters of multicore machines equipped with several Network Interfaces, performance of communication between processes thus depends on which cores these processes are scheduled on, and on their distance to the Network Interface Cards involved. We propose a technique allowing multirail communication between processes to carefully distribute data among the network interfaces so as to counterbalance NUIOA effects. We demonstrate the relevance of our approach by evaluating its implementation within OpenMPI on a Myri-10G + InfiniBand cluster.

  • General Meta Tags

    41
    • title
      Adaptive MPI Multirail Tuning for Non-Uniform Input/Output Access - Inria - Institut national de recherche en sciences et technologies du numérique
    • charset
      UTF-8
    • viewport
      width=device-width, initial-scale=1, shrink-to-fit=no
    • citation_language
      en
    • DC.language
      en
  • Open Graph Meta Tags

    5
    • og:type
      proceedings
    • og:title
      Adaptive MPI Multirail Tuning for Non-Uniform Input/Output Access
    • og:url
      https://inria.hal.science/inria-00486178
    • og:description
      Multicore processors have not only reintroduced Non-Uniform Memory Access (NUMA) architectures in nowadays parallel computers, but they are also responsible for non-uniform access times with respect to Input/Output devices (NUIOA). In clusters of multicore machines equipped with several Network Interfaces, performance of communication between processes thus depends on which cores these processes are scheduled on, and on their distance to the Network Interface Cards involved. We propose a technique allowing multirail communication between processes to carefully distribute data among the network interfaces so as to counterbalance NUIOA effects. We demonstrate the relevance of our approach by evaluating its implementation within OpenMPI on a Myri-10G + InfiniBand cluster.
    • og:url
      https://inria.hal.science/inria-00486178/document
  • Link Tags

    12
    • apple-touch-icon
      /assets/favicon/apple-touch-icon.png
    • dns-prefetch
      //cdn.mathjax.org
    • http://www.w3.org/ns/ldp#inbox
      https://inbox.hal.science/
    • icon
      /assets/favicon/favicon-32x32.png
    • icon
      /assets/favicon/favicon-16x16.png

Emails

1

Links

92