Hot-Spot Avoidance With Multi-Pathing Over Infiniband: An MPI Perspective

PDF Version Also Available for Download.

Description

Large scale InfiniBand clusters are becoming increasingly popular, as reflected by the TOP 500 Supercomputer rankings. At the same time, fat tree has become a popular interconnection topology for these clusters, since it allows multiple paths to be available in between a pair of nodes. However, even with fat tree, hot-spots may occur in the network depending upon the route configuration between end nodes and communication pattern(s) in the application. To make matters worse, the deterministic routing nature of InfiniBand limits the application from effective use of multiple paths transparently and avoid the hot-spots in the network. Simulation based studies ... continued below

Physical Description

PDF-file: 11 pages; size: 0.2 Mbytes

Creation Information

Vishnu, A; Koop, M; Moody, A; Mamidala, A R; Narravula, S & Panda, D K March 6, 2007.

Context

This article is part of the collection entitled: Office of Scientific & Technical Information Technical Reports and was provided by UNT Libraries Government Documents Department to Digital Library, a digital repository hosted by the UNT Libraries. More information about this article can be viewed below.

Who

People and organizations associated with either the creation of this article or its content.

Publisher

Provided By

UNT Libraries Government Documents Department

Serving as both a federal and a state depository library, the UNT Libraries Government Documents Department maintains millions of items in a variety of formats. The department is a member of the FDLP Content Partnerships Program and an Affiliated Archive of the National Archives.

Contact Us

What

Descriptive information to help identify this article. Follow the links below to find similar items on the Digital Library.

Description

Large scale InfiniBand clusters are becoming increasingly popular, as reflected by the TOP 500 Supercomputer rankings. At the same time, fat tree has become a popular interconnection topology for these clusters, since it allows multiple paths to be available in between a pair of nodes. However, even with fat tree, hot-spots may occur in the network depending upon the route configuration between end nodes and communication pattern(s) in the application. To make matters worse, the deterministic routing nature of InfiniBand limits the application from effective use of multiple paths transparently and avoid the hot-spots in the network. Simulation based studies for switches and adapters to implement congestion control have been proposed in the literature. However, these studies have focused on providing congestion control for the communication path, and not on utilizing multiple paths in the network for hot-spot avoidance. In this paper, we design an MPI functionality, which provides hot-spot avoidance for different communications, without a priori knowledge of the pattern. We leverage LMC (LID Mask Count) mechanism of InfiniBand to create multiple paths in the network and present the design issues (scheduling policies, selecting number of paths, scalability aspects) of our design. We implement our design and evaluate it with Pallas collective communication and MPI applications. On an InfiniBand cluster with 48 processes, collective operations like MPI All-to-all Personalized and MPI Reduce Scatter show an improvement of 27% and 19% respectively. Our evaluation with MPI applications like NAS Parallel Benchmarks and PSTSWM on 64 processes shows significant improvement in execution time with this functionality.

Physical Description

PDF-file: 11 pages; size: 0.2 Mbytes

Source

  • Presented at: CCGrid 07 - Seventh IEEE International Symposium on Cluster Computing and the Grid, Rio de Janeiro, Brazil, May 14 - May 17, 2007

Language

Item Type

Identifier

Unique identifying numbers for this article in the Digital Library or other systems.

  • Report No.: UCRL-CONF-228725
  • Grant Number: W-7405-ENG-48
  • Office of Scientific & Technical Information Report Number: 908380
  • Archival Resource Key: ark:/67531/metadc881481

Collections

This article is part of the following collection of related materials.

Office of Scientific & Technical Information Technical Reports

What responsibilities do I have when using this article?

When

Dates and time periods associated with this article.

Creation Date

  • March 6, 2007

Added to The UNT Digital Library

  • Sept. 22, 2016, 2:13 a.m.

Description Last Updated

  • Dec. 5, 2016, 7:52 p.m.

Usage Statistics

When was this article last used?

Congratulations! It looks like you are the first person to view this item online.

Interact With This Article

Here are some suggestions for what to do next.

Start Reading

PDF Version Also Available for Download.

Citations, Rights, Re-Use

Vishnu, A; Koop, M; Moody, A; Mamidala, A R; Narravula, S & Panda, D K. Hot-Spot Avoidance With Multi-Pathing Over Infiniband: An MPI Perspective, article, March 6, 2007; Livermore, California. (digital.library.unt.edu/ark:/67531/metadc881481/: accessed September 26, 2017), University of North Texas Libraries, Digital Library, digital.library.unt.edu; crediting UNT Libraries Government Documents Department.