|Name:||BoF 13: Gearing Up Application Performance for Intel Xeon Phi (KNL)-Based Supercomputers|
|Time:||Wednesday, June 22, 2016
08:30 am - 09:30 am
|Speaker:||Richard Gerber, NERSC|
|Chris Newburn, Intel|
|Thomas Steinke, ZIB|
|Abstract:||This BOF, conducted by the Intel Xeon Phi Users Group (IXPUG) seeks to build community among those developing HPC applications for systems incorporating the Intel Xeon Phi many-core processor. This BOF builds on the successful ones at SC15, ISC2015, and SC14 that drew up to 150 participants each. IXPUG is an independent users group for anyone with in interest in application performance on the Intel Xeon Phi. See http://ixpug.org
Within months after ISC2016, large HPC systems based on the Xeon Phi Knight’s Landing (KNL) processor will make their appearance. Taking advantage of the processor’s full capabilities requires tuning and optimizing code using programming techniques and tools targeted at a combination of CPUs and the Xeon Phi. Threading, vectorization, memory contiguity and alignment, and data locality are important for performance.
The BOF is a place to share experiences and gain insights.
The BOF will start with Lightening Talks from users who have already early-access to KNL, who will share key insights and best practices. That will be followed by a moderated discussion among all those in attendance, which will include representatives from computer and science domains as well as from Intel and other software and tool providers. The BOF will close with an invitation to an ongoing discussion through the Intel Xeon Phi Users Group (IXPUG).
Application developers targeting Intel Xeon Phi (KNL) supercomputers, experts in code optimization for Xeon Phi (KNL) platform, performance tools users and developers.