SC13 Home > SC13 Schedule > SC13 Presentation - A Transactional Model for Fault-Tolerant MPI for Petascale and Exascale Systems

SCHEDULE: NOV 16-22, 2013

When viewing the Technical Program schedule, on the far righthand side is a column labeled "PLANNER." Use this planner to build your own schedule. Once you select an event and want to add it to your personal schedule, just click on the calendar icon of your choice (outlook calendar, ical calendar or google calendar) and that event will be stored there. As you select events in this manner, you will have your own schedule to guide you through the week.

A Transactional Model for Fault-Tolerant MPI for Petascale and Exascale Systems

SESSION: Research Poster Reception

EVENT TYPE: Posters, Electronic Posters, and Education Posters

TIME: 5:15PM - 7:00PM

AUTHOR(S):Amin Hassani, Anthony Skjellum, Ron Brightwell

ROOM:Mile High Pre-Function

ABSTRACT:
Fault-Aware MPI (FA-MPI) is a novel approach to provide fault-tolerance through a set of extensions to the MPI Standard. It employs a transactional model to address failure detection, isolation, mitigation, and recovery via application-driven policies. This approach allows applications to employ different fault-tolerance techniques, such as algorithm-based fault tolerance (ABFT) and multi-level checkpoint/restart methods. The goal of FA-MPI is to support fault-awareness in MPI objects and enable applications to run to completion with higher probability than running on a non-fault-aware MPI. FA-MPI leverages non-blocking communication operations combined with a set of TryBlock API extensions that can be nested to support multi-level failure detection and recovery. Managing fault-free overhead is a key concern as well.

Chair/Author Details:

Amin Hassani - University of Alabama at Birmingham

Anthony Skjellum - University of Alabama at Birmingham

Ron Brightwell - Sandia National Laboratories

Add to iCal  Click here to download .ics calendar file

Add to Outlook  Click here to download .vcs calendar file

Add to Google Calendarss  Click here to add event to your Google Calendar