The Argonne Training Program on Extreme-Scale Computing starts next week. Paul Messina believes more institutions - including universities - need to offer such advanced training. The complexities of exascale computing are the reason as programming the current generation of leadership class machines requires knowledge of a broad spectrum of topics and many skills: both … [Read more...]
SC14 Technical Program and Registration – XSEDE/TACC Resources for Farber Tutorial
Register early for Supercomputing 2014 in New Orleans and save up to $275. View the Technical Program online (and register for our tutorial!) The Technical Program fee includes admission to all conference sessions, exhibits, the Monday night Exhibits opening event, Thursday night event, and one copy of the SC14 proceedings. Click here to view the grid showing access to … [Read more...]
Intel Webinar About “Omni Scale” Next-Generation High-Performance Fabric and Future Directions
Register to attend the August 5th webinar by Joe Yaworski from the Intel Technical Computing Group titled, "High Performance Fabrics from Intel - Today and Tomorrow". Intel's thinking about network infrastructure will have a decided impact on the future of HPC given their recent wins on the NNSA (National Nuclear Safety Agency) 42 PF/s Trinity supercomputer plus the NERSC … [Read more...]
Model Validation: Darth Vader for President
Too good to pass up (and never forget the importance of validation!): The website FiveThirtyEight (a wonderful statistical analysis news site) reported that the data from their article Luke Skywalker vs. Hillary Clinton was further analyzed by Christopher Ingraham at the Wonkblog, 'Walter Hickey on the popularity of “Star Wars” characters' with the addition … [Read more...]
Linus Torvalds Says Fix GCC 4.9 Code Generation!
Phoronix picked up Linus Torvalds' providing some not so gentle feedback on GCC 4.9. GCC 4.9 supports OpenMP 4.0. Apparently the latest GNU compiler is doing some silly spilling of CPU registers (including constants!) that caused a random panic in a load balance function with the in-development Linux 3.16 kernel. On a comparative note, GCC just received … [Read more...]
42 PF/s Trinity Supercomputer to Use Intel Knights Landing
First details on the National Nuclear Security Administration (NNSA) Trinity Supercomputer show that the 42 PF/s system costing $174M USD will run a combination of Intel Haswell and Knights Landing processors. In particular the Intel Xeon Phi devices will use Micron’s Hybrid Memory Cube technology, which will greatly help memory bandwidth and memory capacity limited … [Read more...]
TechEnablement Becomes an SC14 Media Partner
TechEnablement.com is growing! We are proud to announce that we are now a Supercomputing 2014 (SC14) Media Partner. SC14 partners with media organizations covering high performance computing, networking, storage and analysis. Such partnerships are established with the goal of disseminating news about the conference to key audiences while helping the media organizations raise … [Read more...]
NVIDIA Launches Women and CUDA Site
Today NVIDIA launched the new Women and CUDA website, representing female students, professors and industry experts from around the world. Take a look. This follows up on NVIDIA's Women@GTC event in March that invited women developers, researchers and scientists to share how they utilize GPU computing in their work. TechEnablement previous reported on women in technology events … [Read more...]
Part 2: No Idle Time CUDA Task Parallelism Across Eight GPUs
Part 1 in this tutorial series showed that task-based parallelism using concurrent kernels can accelerate applications simply by plugging more GPUs into a system - just as the GPU strong scaling execution model can accelerate applications simply by installing a newer GPU containing more SMX (Streaming Multiprocessors). No recompilation required! NVIDIA nvvp timelines show very … [Read more...]
NCSA (XSEDE) to Host OpenACC Aug 5th Workshop Using Blue Waters – Only Few Sites Can Receive Telecast
NCSA and the Extreme Science and Engineering Discovery Environment (XSEDE) project that it leads, along with the Pittsburgh Supercomputing Center, will present an OpenACC GPU programming workshop on Aug. 5. The workshop will have a hands-on component using NCSA’s Blue Waters supercomputer. Register through the online XSEDE portal. Send questions to Tom Maiden at … [Read more...]