HPCwire: Forum Adopts MPI 2.1 Standard

Since 1986 - Covering the Fastest Computers
in the World and the People Who Run Them

Translation Disclaimer

Subscribe | Sign In

Visit additional Tabor Communication Publications

Leading HPC
Solution Providers

September 24, 2008

Forum Adopts MPI 2.1 Standard

Sept. 24 -- A consortium of computing experts has approved the first upgrade to the venerable Message Passing Interface (MPI) standard since 1997.

Ratified September 4 in Dublin, Ireland, MPI version 2.1 introduces modest changes to the standard in anticipation of more ambitious revisions expected in the next couple of years. According to MPI Forum Chairman Rich Graham of Oak Ridge National Laboratory's National Center for Computational Sciences, the 600-page MPI 2.1 is primarily an attempt to clarify and correct errors in the 11-year-old standard.

"The 2.0 document has errors in it -- typos and the like," Graham noted. "It's also not a single document. You sometimes have to look at several locations in the standard to find everything it says about a particular topic, so we made an attempt to bring those together so it's easier to find."

MPI, which allows applications to run on parallel computers, is used by nearly all scientific application developers to exchange data between processors. Graham credited Rolf Rabenseifner at the University of Stuttgart's High Performance Computing Center Stuttgart for spearheading the MPI 2.1 effort.

The next round of changes to the standard will appear in version 2.2, with that effort spearheaded by William Gropp of the University of Illinois at Urbana-Champaign's National Center for Supercomputing Applications. Graham hopes the forum will decide on textual changes and application performance interfaces by the end of the year, with final ratification in mid-2009. He noted that those changes will also be relatively minor and easy to implement.

More fundamental issues will be addressed in MPI 3.0, which the group hopes to ratify by the end of 2010. That standard will tackle a variety of issues, including remote memory access (the ability for one processor to write to another processor's memory), fault tolerance (the ability to respond to a problem without crashing the application), and non-blocking collectives (methods for simplifying and improving communication within an application).

Graham stressed that versions 2.1 and 2.2 are not expected to break existing applications that use the standard. While the more ambitious 3.0 standard may break existing applications by deprecating some functions found in 2.2, he noted that backward compatibility will be maintained for the foreseeable future.

The complete MPI 2.1 standard and information for getting involved in this effort are available at the MPI Forum Web site (www.mpi-forum.org).

-----

Source: Oak Ridge National Laboratory

Share Options

Subscribe

Subscribe to HPCwire

Discussion

There are 0 discussion items posted.

Join the Discussion

Become a Registered User Today!

Registered Users Log in join the Discussion

RSS Feeds

Feeds by Topic

Feeds by Industry

Feeds by Content Type

Subscribe to All Content

All

May 23, 2012

May 22, 2012

May 21, 2012

May 18, 2012

May 17, 2012

May 16, 2012

May 15, 2012

May 14, 2012

May 11, 2012

Feature Articles

NVIDIA Works On CPU Co-Dependency Issues with Kepler GPU

NVIDIA is telling everyone that the GK110, its new Kepler GPU aimed at supercomputing, is all about improving performance per watt. But the other driving theme behind the new architecture is reducing the GPU's reliance on its CPU host. How well it accomplishes both these goals areas could determine the success of the new chip in high performance computing.
Read more...

OpenACC Starts to Gather Developer Mindshare

PGI, Cray, and CAPS enterprise are moving quickly to get their new OpenACC-supported compilers into the hands of GPGPU developers. At NVIDIA's GPU Technology Conference this week, there was plenty of discussion around the new HPC accelerator framework, and all three OpenACC compiler makers, as well as NVIDIA, were talking up the technology.
Read more...

NVIDIA Launches Kepler Into HPC

NVIDIA has introduced its first Kepler-generation GPU product for high performance computing, and revealed some of the inner working of the new architecture. The announcement took place at the kickoff of the company's GPU Technology Conference taking place this week in San Jose, California.
Read more...