“The promising new parameter in place of the transistor count is the perceived increase in the capacity and bandwidth of storage, driven by device, architectural, as well as packaging innovations: DRAM-alternative Non-Volatile Memory (NVM) devices, 3-D memory and logic stacking evolving from VIAs to direct silicone stacking, as well as next-generation terabit optics and networks. The overall effect of this is that, the trend to increase the computational intensity as advocated today will no longer result in performance increase, but rather, exploiting the memory and bandwidth capacities will instead be the right methodology.”
Today AMD, ARM, Huawei, IBM, Mellanox, Qualcomm, and Xilinx announced a collaboration to bring the CCIX high-performance open acceleration framework to data centers. The companies are collaborating on the specification for the new Cache Coherent Interconnect for Accelerators (CCIX). For the first time in the industry, a single interconnect technology specification will ensure that processors using different instruction set architectures (ISA) can coherently share data with accelerators and enable efficient heterogeneous computing – significantly improving compute efficiency for servers running data center workloads.
Editor’s Note: The OpenPOWER European Summit has been officially postponed until later in 2016, but don’t miss the International Workshop on OpenPOWER for HPC on June 23 at ISC 2016. The OpenPOWER Foundation was established as a non-profit consortium to give its members the ability to innovate software/hardware solutions based on the POWER architecture. About half of […]
University of Texas at Arlington physicists are preparing the Titan supercomputer at Oak Ridge Leadership Computing Facility in Tennessee to support the analysis of data generated from the quadrillions of proton collisions expected during this season’s Large Hadron Collider particle physics experiments.
Today the HPC Advisory Council announced its Fourth Annual RDMA Programming Competition in China. Designed to support undergraduate curriculum and talent development, this unique hands-on competition furthers students study, experience and mastery.
Today the ISC Group announced the appointment of Prof. Dr. Jack Dongarra of University of Tennessee, USA, as the program chairman for ISC 2017. This appointment underlines the international dimension of this conference. “I am honored to be named the program chair for ISC High Performance in 2017. I have been participating and actively contributing to ISC every year over the last three decades, first as a speaker, and later as a collaborator with Hans Meuer, Erich Strohmaier, and Horst Simon as one of the TOP500 authors,” said Dongarra. “My mission as program chair for ISC 2017 is to continue to improve the quality of the ISC technical program and to help expand the HPC community.”
In this special guest feature, Lance Farrell from Science Node writes that CIPRES is a web-based gateway that allows researchers to easily explore evolutionary relationships between species using NSF supercomputers. “In the six years since it was established, CIPRES has enabled 2,300+ scientific publications, while only occupying about 1% of the NSF supercomputing resources. That’s an impressive return on investment.”
In this video, ITIF hosts a hearing on the The Vital Importance of High-Performance Computing to U.S. Competitiveness and National Security. Their recently published report urges U.S. policymakers to take decisive steps to ensure the United States continues to be a world leader in high-performance computing.
In this TACC podcast, Ari Kahn from the Texas Advanced Computing Center and Eddie Garcia from Cloudera describe a recent Hackathon in Austin designed to tackle data challenges in the fight against the Zika virus. The Texas Advanced Computing Center provided time on the Wrangler data intensive supercomputer as a virtual workspace for the Zika hackers.
“With NVIDIA GPU technology on IBM Cloud, we are one step closer to offering supercomputing performance on a pay-as-you-go basis, which makes this new approach to tackling big data problems accessible to customers of all sizes,” says Jerry Gutierrez, HPC leader for SoftLayer, an IBM Company. “We’re at an inflection point in our industry, where GPU technology is opening the door for the next wave of breakthroughs across multiple industries.”