High Performance Computing Networking

Today, a small group of us left “stealth mode” and presented a group of ideas to the MPI Forum loosely grouped under the moniker of “MPI Sessions.” These ideas behind this presentation originated from several places: My previous presentation of crazy ideas about INIT/FINALIZE at EuroMPI’15 in Bordeaux, France That Jonathan Dursi blog entry Other […]

Here’s the slides from the Open MPI State of the Union (SotU) BOF that were presented by George Bosilca, Nathan Hjelm, and myself at the SC’15 trade show. Thanks to all who were able to join us at the BOF in person! UPDATE: Per request, I have also added the slides from the MPI Forum […]

Who’s heading to SC’15 this weekend?  I am! Austin looks to have an exciting lineup this year.  Here’s the things I’m involved with: Dave Goodell and the rest of the Libfabric crew will be presenting An Introduction to the OpenFabric Interfaces API tutorial Dr. George Bosilca and I will presenting the annual Open MPI State […]

I had a good conversation with an ISV yesterday who makes a popular MPI-based simulation application.  One of the things I like to do in these kinds of conversations is ask the ISV engineers two questions: What new features do you want from the MPI implementations that you use? What new features or changes do you want […]

In the previous blog entry, I shared the slides of one of the mini-talks that I gave at EuroMPI 2015 in Bordeaux, France (and don’t forget to start planning for EuroMPI 2016 in Edinburgh, Scottland). The second mini-talk I gave was twofold in itself: I discussed Cisco’s journey from the Verbs API to the Libfabric API, and […]

I recently had the pleasure of attending EuroMPI 2015, hosted by INRIA in Bordeaux, France (…hey, you should attend EuroMPI 2016 in Edinburgh, Scottland!). I gave two mini-talks during my speaking slot, the first of which was entitled: Crazy ideas about revamping MPI_INIT and MPI_FINALIZE.

I recently received an interesting email from Heiko Bauke about a new C++-based message-passing library that he is working on called MPL. His library aims to make a simple-to-use library that exploits the features available in modern C++ compliers — it’s not a simple 1:1 mapping of C-to-C++ bindings like the MPI 2.x C++ bindings were.

Similar to previous versions of the standard, the MPI-3.1 standard is now available in hardcover. Even though I have a copy of the MPI-3.1 PDF for quick/easy reference on my laptop, I find something intangibly inherently useful (and comforting?) in having a paper copy of the standard to thumb through. The books have been (literally) […]

Later today, Sean Hefty will present a paper about the OpenFabrics Interfaces (a.k.a. “libfabric“) at the 2015 IEEE Hot Interconnects conference. Libfabric is the next-generation Linux library being developed by an open source consortium of vendors and academic researchers that implements the OpenFabrics Interfaces, specifically designed to expose application-focused networking functionality to high performance applications (e.g., MPI, PGAS, […]