Sign up for our newsletter and get the latest HPC news and analysis.
Send me information from insideHPC:


NEC Supercomputer at JGU in Germany Ranks #65 on TOP500

NEC Deutschland GmbH has delivered an LX series supercomputer to Johannes Gutenberg University Mainz (JGU), one of Germany’s leading research universities and part of the German Gauss Alliance consortium of excellence in high-performance computing. The new HPC cluster ranks 65th in the most current TOP500 list of the fastest supercomputers in the world from November 2017 and 51st in the Green500 list of the most energy-efficient supercomputers.

Hyperion Innovation Excellence Award goes to UberCloud and Stanford Living Heart Project

Last week at SC17, Hyperion Research announced that the UberCloud and the Stanford Living Heart Project have won the Hyperion Award for Innovation Excellence. “The Stanford LHP project is simulating cardiac arrhythmia, which can be an undesirable and potentially lethal side effect of drugs. The electrical activity of the heart turns chaotic, decimating its pumping function, thus diminishing the circulation of blood through the body.”

Russian RSC Group Joins the Intel Select Solutions for HPC Program

Last week at SC17, Russian HPC vendor RSC Group showcased the next generation of their liquid-cooled RSC Tornado supercomputers based on Intel Xeon Scalable Processors. “RSC is demonstrating a full set of components for modern HPC computing systems of different scale with 100% direct liquid cooling in ‘hot water’ mode, including high-performance RSC Tornado computing nodes based on the top-bin Intel Xeon Platinum and Intel Xeon Gold processors (part of the Intel Xeon Scalable platform), Intel Server Board S2600BP, high-speed NVMe solid state drives in high-dense М.2 format and the latest Intel Optane SSD DC P4800X Series. The RSC Tornado solution is built using 100% direct liquid cooled Intel Omni-Path Edge Switch 100 Series that ensures end-to-end efficiency of the cooling solution with ‘hot water’ and eventually the lowest possible total cost of ownership of the system.”

Video: An Affordable Supercomputing Testbed based on Rasberry Pi

In this video from SC17, Bruce Tulloch from BitScope describes a low-cost Rasberry Pi cluster that LANL can use to simulate large-scale supercomputers. “The BitScope Pi Cluster Modules system creates an affordable, scalable, highly parallel testbed for high-performance-computing system-software developers. The system comprises five rack-mounted BitScope Pi Cluster Modules consisting of 3,000 cores using Raspberry Pi ARM processor boards, fully integrated with network switching infrastructure.”

Dell EMC Brings Machine Learning to Mainstream Enterprises

This week at SC17, Dell EMC announced new machine learning and deep learning solutions, continuing the company’s work to bring high performance computing and data analytics capabilities to mainstream enterprises worldwide. “When you think about what this means for industries like financial services or personalized medicine, the possibilities are endless and exciting.”

New TeraBox 1U FPGA Server Doubles Compute and I/O Density

Today BittWare announced the TeraBox 1432D high density 1U FPGA server at SC17. Based on a Dell PowerEdge C4130, the customized TeraBox 1432D provides an unprecedented thirty-two 100GbE ports directly connected to four large Xilinx or Intel FPGAs on BittWare accelerator cards. This server targets users requiring the highest density of large FPGAs with directly-connected I/O for clustering or networking.

Video: Atos and ParTec to deploy 12 Petaflop Supercomputer at Jülich

In this video, Hugo Falter from Par-Tec describes the new 12 Petaflop supercomputer coming to the Jülich Supercomputing Centre in Germany. “Modular supercomputing, an idea conceived by Dr. Lippert almost 20 years ago, was realised by JSC and ParTec in the EU-funded research projects DEEP and DEEP-ER together with many partners from research and industry. Since 2010, our experts have been developing the software, which will in future create the union of several modules into a single system.”

BOXX rolls out AMD EPYC Deep Learning Server at SC17

This week at SC17, BOXX Technologies debuted the new GX8-M server, featuring dual AMD EPYC 7000-series processors, eight full-size AMD or NVIDIA graphics cards, and other innovative features designed to accelerate high performance computing applications. “BOXX is taking the lead with deep learning solutions like the GX8-M which enables users to boost high performance computing application performance and accelerate their workflows like never before.”

Lenovo Accelerates AI to Solve Humanity’s Greatest Challenges

Today at SC17, Lenovo announced new initiatives designed to empower customers to embrace Artificial Intelligence and make it a true reality for their organizations to achieve augmented intelligence capabilities for increased productivity and transformative results. “Artificial intelligence is already having a profound impact on traditional business strategies and scientific research, and most senior leaders consider it a priority for the year ahead. To truly benefit from the vast amount of data available to organizations today, our customers must embrace AI as the vehicle to help them achieve success in today’s competitive business landscape,” said Kirk Skaugen, president, Lenovo Data Center Group. “With our newly opened, global AI innovation centers and a comprehensive product and service portfolio we are committed to helping bring their AI deployments to life.”

Intel Select Solutions for HPC Debut at SC17

Intel Select Solutions HPC provide a fast path for purchasing and deploying a cluster for simulation and modeling workloads with a pre-validated selection of components designed to meet the demands of HPC applications and workflows. These systems provide the capabilities and agility needed to support a range of different workloads and reduce or eliminate the need for multiple single-purpose systems. In addition, the performance of key system characteristics are verified for Intel Select Solutions for Simulation and Modeling at both the node and cluster level.