titan

Up since 11/8/17 02:45 pm

eos

Up since 11/14/17 11:20 pm

rhea

Up since 10/17/17 05:40 pm

hpss

Up since 11/20/17 09:15 am

atlas1

Up since 11/15/17 07:25 am

atlas2

Up since 11/27/17 10:45 am
OLCF User Assistance Center

Can't find the information you need below? Need advice from a real person? We're here to help.

OLCF support consultants are available to respond to your emails and phone calls from 9:00 a.m. to 5:00 p.m. EST, Monday through Friday, exclusive of holidays. Emails received outside of regular support hours will be addressed the next business day.

Loading Events

A Preview of MPI 3.0: The Shape of Things to Come

Details

Start:
June 25, 2012
End:
June 28, 2012

Venue

ORNL Wigner Auditorium (B:4500N)
1 Bethel Valley Road
Oak Ridge, 37831 United States

Organizer

Manjunatha Gorentla Venkata
Phone:
(865) 574-5949
Email:
manjugv@ornl.gov

Abstract

Abstract

The Message Passing Interface (MPI) standardization body is preparing to release version 3.0 of the MPI standard later this year. The MPI standard is the predominate message passing programming interface used by most High Performance Computing (HPC) applications. The MPI 3.0 release focuses on scalability and programability advancements derived from years of research into HPC applications and systems. This release contains features poised to have a significant impact on scientific HPC application development such as nonblocking collectives, improved one-sided communication, and new FORTRAN interfaces. In preparation for the MPI 3.0 release, this seminar series will present an overview of new MPI standard directed at HPC application developers. The seminar series will introduce the new functionality, discuss potential application benefits, and provide information about early access to prototype MPI implementations. In addition to our discussion of MPI 3.0, we will also highlight new features of the current MPI 2.2 standard, and review developing proposals targeted at the next MPI standard. Application developers are encouraged to highlight the most important features they would need for their applications which will serve to prioritize research efforts at ORNL.

Manju Gorentla Venkata and Josh Hursey
Application Performance Tools Group, ORNL
Computer Science and Mathematics Division Seminar
3:00 PM — 4:00 PM, Building 4500-N, Wigner Auditorium
Contact: Cindy Sonewald (sonewaldc@ornl.gov), 865.574.3125

Agenda

Monday, June 25, 2012

Part 1 Presentation Speaker
MPI Process (brief)
Timeline to 3.0
MPI 3.0 Fortran Bindings
Joshua Hursey
Part 1b MPI 2.2 Manjunath Gorentla Venkata

Tuesday, June 26, 2012

Part 2 Presentation Speaker
Collectives
-Neighborhood
-Nonblocking
Communicator Creation
-Noncollective
-Nonblocking Duplication
Manjunath Gorentla Venkata

Thursday, June 28, 2012

Part 3 Presentation Speaker
MPI Matched Prob/Recv
RMA/One-sided Enhancements
Tool Interfaces
MPI (next)
-Fault Tolerance
-Hybrid, Collectives, …
Joshua Hursey