Sign up for our newsletter and get the latest HPC news and analysis.

Search Results for: mpi

It’s Time for MPI Over Everything (MPIoE)

jeff

Over at Google Plus, Jeff Squyres has posted a somewhat belated April 1 post announcing an exciting new technology called the MPI over Everything Project (MPIoE). “One key innovative technology developed as part of the MPIoE effort is the Far Reaching Internet Datagram Efficiency (FRIDGE) framing protocol.”

Open MPI 1.7.4 Goes Fortran-Ready

Over at The Register, Richard Chirgwin writes that the decade-old Open-MPI project has shipped code for a major feature release that brings it close to a complete MPI 3 implementation.

Online Tutorial: Hybrid MPI and OpenMP Parallel Programming

openmp

OpenMP.org has posted the slides and audio from a day-long tutorial on Hybrid MPI and OpenMP Parallel Programming from SC13.

Slidecast: MPI Requirements of the Network Layer

jeff

Jeff Squyres from Cisco describes the proposed successor to the Linux verbs API that is designed to better serve the needs of MPI. “It’s not libibverbs 2.0 — it’s a new API that aims to both expand the scope of what libibverbs did, and also to address many of its much-criticized shortcomings.”

A Gentle Introduction to MPI

prace

PRACE has developed an extensive series of online HPC Tutorials. In this video, CSC provides a basic introduction to parallel programming concepts such as task/data parallelism and an introduction to parallel programming concepts such as parallel scaling and Amdahl’s law.

Scalable and Resilient MPI and GPUDirect RDMA

panda

In this video from the Mellanox booth at SC13, DK Panda from Ohio State presents: High Performance Scalable and Fault Resilient MPI and GPUDirect RDMA. See more talks from the Mellanox Booth.

A New MPICH ABI Compatibility Initiative

As announced at SC13, the producers of several notable MPICH-derived MPI implementations have begun a collaboration with the explicit goal of maintaining ABI compatibility between their implementations. Without such compatibility between implementations,” said Kenneth Raffenetti, a software developer in Argonne’s Mathematics and Computer Science Division, “every new release would require application developers to rebuild and […]

Learn to Redefine Scalable OpenMP and MPI Price-to-Performance

Learn to redefine your HPC investment by delivering all necessary compute capabilities while enjoying the ease of scalable shared memory computing!

Cray Provides Cray Compiler Environment (CCE) to Its Line of Cluster Supercomputers

images

At SC13, Cray announced that the latest release of the Cray Compiler Environment (CCE) is now available on the Cray CS300 line of cluster supercomputers.

Power Savings – the Quest for Green MPI

dellslide

This presentation investigates whether power savings on the processor and system level can be obtained while an MPI rank is waiting (blocking) for incoming data.