Aug 18, 2015 – The OpenMP Architecture Review Board (ARB) today announced the abstract of presentations to be hosted at the inaugural OpenMPCon Developer Conference, which will be held in Aachen, Germany, 28-30 September, 2015. For full details and registration visit: www.openmpcon.org
The latest hot topics on exascale and heterogeneous computing will be discussed in a comparison of the latest state of OpenMP versus OpenACC that will be offered by James Beyers from NVIDIA. As both an OpenMP and OpenACC insider, he will present his opinion of the current status of these two directive sets for programming “accelerators”. Insights into why some decisions were made and why they were changed later will be used to explain the trade-offs required to achieve agreement on these complex directive sets.
One of the main aims of OpenMPCon is for users and developers to learn the latest tips, and tricks from OpenMP Gurus. There is none better than the past OpenMP Language Chair Mark Bull from EPCC. This talk will present a series of practical hints for OpenMP programmers, collected from many years experience of teaching OpenMP, and answering questions on the OpenMP Forum.
If a user wants performance in OpenMP, then he will need to attend the talk on where did your performance go. Mark Bull will give an understanding of how and why OpenMP programs lose performance. In this talk he will attempt to enumerate all the possible ways that OpenMP programs can deliver less than ideal speedup, namely lack of parallelism, load imbalance, synchronization, communication, hardware contention and compiler non-optimization.
OpenMPCon is partnering closely with IWOMP 2015, the research focused International Workshop on OpenMP which will take place immediately after OpenMPCon. Developers are encouraged to attend both events.
Sponsors for OpenMPCon include: Appentra, cOMPunity, hgpu.org, insideHPC, Intel, JARA, OpenMP ARB, RWTH Aachen University, Silexica and Texas Instruments.
About OpenMP ARB
The OpenMP ARB has as mission to standardize directive-based multi-language high-level parallelism that is performant, productive and portable. Jointly defined by a group of major computer hardware and software vendors and major parallel computing user facilities, the OpenMP API is a portable, scalable model that gives parallel programmers a simple and flexible interface for developing parallel applications for platforms ranging from embedded systems and accelerator devices to multicore systems and shared-memory systems. The OpenMP ARB owns the OpenMP brand, oversees the OpenMP specification and produces and approves new versions of the specification.