 |
Subscribe |  |
Share |
At ISC 2011, NVIDIA and a large number of partners, comprising the GPU Computing ecosystem will be featuring NVIDIA® GPU computing solutions that are powering the next wave in HPC. We invite you to attend the featured ISC sessions, sign up for an advanced CUDA tutorial and visit us (booth #630) and our partners to learn more about the latest NVIDIA Tesla™ GPU computing solutions. 
HIGHLIGHTS
DebatE: GPUs – The Fast Lane on the Road to Better Science? Tuesday, June 21, 2011 | 11:30am - 1:00pm | Hall 3 Chair: Horst Simon, Lawrence Berkeley National LaboratoryContributors: Thomas Sterling, Louisiana State University David Kirk, NVIDIA NVIDIA SESSIONS
SUNDAY, JUNE 19, 2011 | 9AM - 6PM | HAMBURG UNIVERSITY, MAIN BUILDING, ROOM A PDF PRESENTATIONS AVAILABLE FOR DOWNLOAD BELOW CUDA is a general-purpose architecture for writing highly parallel applications. In this CUDA tutorial, NVIDIA engineers will partner with academic and industrial researchers to present CUDA and discuss its advanced use for science and engineering domains.
BOF #7: ABOUT DATA-PARALLELISM / SIMD FOR HPC IN THE FUTURE WEDNESDAY, JUNE 22, 2011 | 2:30PM - 3:15PM | HALL B In order for HPC users to perform more computation-intensive tasks, they must take full advantage of data-parallelism in order to exploit parallel hardware. SIMD instructions supports data parallelism yet software that are compute-intensive fail to take full advantage of the instructions. Has the HPC software and application industry forgotten how to utilize vector/SIMD processing? Join this BOF as we look at new software concepts that leverage and exploit vectorization and SIMD as well as discussing approaches to utilize and extract best performance from modern hardware technologies. BOF #12: GPUs: TRENDING TECHNOLOGY, OR SERIOUS COMPUTING? WEDNESDAY, June 22, 2011 | 2:30PM - 4:45PM | HALL C2.2 To GPU or not to GPU? This BOF invites participants to explore and debate the merits of GPU computing, including how to best maximize their value in computing and how to effectively manage complex clusters comprising both traditional CPUs and GPUs. HOT SEAT SESSION: THE FUTURE OF GPU COMPUTING Wednesday, June 22, 2011 | 3PM – 3:15PM | hall 3 GPUs are playing an increasingly pivotal role in the scientific computing space and are at the forefront of one of the most significant technology transitions in supercomputing since the shift from vector to cluster–based processing. Today, GPUs are enabling scientists and researchers from leading institutions increase the pace and scope of their work, and making discoveries faster than ever before. This presentation will outline HPC industry's move to parallel computing with GPUs, and the future of GPU computing. EXHIBITOR'S FORUM: BUILDING EXASCALE GPU-BASED SUPERCOMPUTERS Wednesday, june 22, 2011 | 10:40AM – 11:10 AM | Exhibition hall, #950 Heterogeneous computing is quickly establishing itself as one of the leading approaches to build exascale supercomputers, while staying within power and monetary budgets. Application developers have to parallelize their applications – whether the system is a heterogeneous system or even a traditional multi-core CPU system. Given these challenges, the opportunities that GPUs represent are overwhelming. Already, there are several large supercomputers built using GPUs in Asia (including the world's fastest Supercomputer Tianhe-1A) and many more coming soon. We will discuss these topics and also talk about programmability and developer productivity. OTHER SESSIONS OF INTEREST
TUTORIAL: "LINPACK ON FUTURE MANYCORE & GPU BASED SYSTEMS" (FEATURING JACK DONGARRA) SUNDAY, JUNE 19, 2011 | 9AM - 1PM | HAMBURG UNIVERSITY, MAIN BUILDING, ROOM M This tutorial will provide a brief historical look at the development of dense linear algebra libraries, from LINPACK, to LAPACK, to ScaLAPACK. Today we see new computer architectures emerging, which will cause another shift in the software landscape, namely multicore and accelerators. These changes will necessitate innovation again in the linear algebra libraries. We have been developing two packages, PLASMA and MAGMA, for just these architectures. The tutorial presents PLASMA (Parallel Linear Algebra Software for Multicore Architectures) from the user perspective, software contributor perspective and core developer perspective. A wide array of topics is covered, ranging from simple usage examples to intricacies of runtime task scheduling. FRIDAY-SATURDAY, JUNE 17-18, 2011 | RADISSON BLUE HOTEL / CONGRESS CENTRE HAMBURG At HP-CAST 16 a special focus will be on the following key topics:
Furthermore numerous tutorials and other in-depth sessions (SIGs) will address specific HPC topics of interest. To register, click here. NVIDIA GPU COMPUTING ECOSYSTEM AT ISC 2011 |
