The BoF will support SC-14 attendees interested in Intel Xeon Phi tuning and debugging, and the IXPUG User Group. A User Group discussion will focus on building a community among those developing HPC applications for systems incorporating the Intel Xeon Phi many-core processor.
|Time:||Wednesday, November 19, 5:30PM – 7:00PM Central Time|
|Location:||Room 286-87 at SC14 in New Orleans, LA|
|Session Leaders:||Richard Gerber, Kent Milfeld, Chris Gottbrath, Chris J. Newburn|
The BoF (Birds of a Feather) will consist of Lightning Talks, an "Open Mic" panel discussion, and a short User Group meeting.
The BoF will start with brief presentations from academia, national labs and industry representatives sharing their experiences. The emphasis of the BoF is on performance tuning and debugging on Phi systems.
Taking advantage of the processor's full capabilities requires tuning and optimizing using programming techniques and tools targeted at a combination of CPUs and the Xeon Phi Coprocessor. Even applications that are already multi-threaded may need to be reworked to expose additional concurrency. Within a manycore system, threading, vectorization, memory contiguity and alignment, and data locality are critical components of performance tuning. Speakers and BoF participants will share insights and experiences with performance tuning and debugging on these systems.
After the Lightning Talks, the speakers and a few Intel experts will convene as a panel for an Open MIC discussion.
Also, TACC, NERSC and LANL will lead a discussion with attendees to plan future meetings and activities of the User Group. The Intel Xeon Phi User's Group (IXPUG), based at TACC, is expanding world-wide to include all HPC sites that incorporate the Phi, whose growing importance in HPC is evidenced by the two Phi-based systems in the top 10 of the June 2014 Top 500. NERSC and LANL recently announced large procurements of next-generation Phi systems.
In addition to user group planning, the work product output of this BoF will be a summary of the talks, with detailed notes on points brought up during the BoF, and the results of a five-question survey taken during the BoF to gather feedback and user experiences beyond those which were covered in the BoF.
The BoF organizers, Richard Gerber of NERSC, Kent Milfeld of TACC, Chris Gottbrath of Rogue Wave, and Chris J. Newburn of Intel will kick off the BoF, moderate presentations and discussion, and frame follow up activities.
(5:30PM, Wednesday, November 19, 2014)
Click on the Title to see slides and notes of the Lightning Talk (LT).
|5:30||-||Welcome/Overview/Introductions||Chaired by: Richard Gerber, Kent Milfeld, Chris Gottbrath, CJ Newburn|
|5:40||LT Talk:||Nambiar, Manoj (TCS)||AMBER –optimizations on Xeon Phi|
|5:45||LT Talk:||Brook, Glenn (JICS)||HPC-BLAST: Scaling the Life Sciences for the Intel® Many Integrated Core future|
|5:50||LT Talk:||Noack, Matthias (ZIB)||Hierarchical Equations of Motion: -What we can learn from OpenCL-|
|5:55||LT Talk:||Chow, Edmond (GA Tech)||Large-Scale Hydrodynamic Brownian Simulations on Intel Xeon Phi|
|6:00||LT Talk:||Poulsen, Jacob Weismann (DMI)||Refactoring for Xeon Phi|
|6:05||LT Talk:||Luszczek, Piotr (UTK)||MAGMA MIC: HPC Linear Algebra for Intel Xeon Phi|
|6:15||-||Open Mic Panel Discussion||-|
|6:45||-||IXPUG Discussion||Draft Charter|
Lightning Talks and Additional Selected Submissions
- Kent Milfeld, TACC