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


Ayar Labs Joins DARPA PIPES Project as Intel Optical IO Provider

Optical startup Ayar Labs has been selected as Intel’s optical I/O solution partner for their recently awarded DARPA PIPES research project. “The goal of PIPES (Photonics in Package for Extreme Scalability) is to develop integrated optical I/O solutions co-packaged with next generation FPGA/CPU/GPU and accelerators in Multi-Chip Packages (MCP) to provide extreme data rates (input/output) at ultra-low power over much longer distances than supported by current technology. In the first phase of the project, the Ayar Labs TeraPHY chiplet will be co-packaged with an Intel FPGA using the AIB (Advanced Interconnect Bus) interface and Intel’s EMIB silicon-bridge packaging.”

How Deep Learning is Driving New Science

In this special guest feature, Robert Roe from Scientific Computing World looks at the development of deep learning and its impact on scientific applications. “In general, it depends on the use case but you can think of two cases where AI is useful. The first case is to solve problems that are hard to solve in a rule-based way, which is a similar domain as you may have outside science for speech recognition or image recognition.”

NVIDIA Announces GPU-Accelerated Supercomputer on Azure

At SC19, NVIDIA announced the availability of a new kind of GPU-accelerated supercomputer in the cloud on Microsoft Azure. “Built to handle the most demanding AI and high performance computing applications, the largest deployments of Azure’s new NDv2 instance rank among the world’s fastest supercomputers, offering up to 800 NVIDIA V100 Tensor Core GPUs interconnected on a single Mellanox InfiniBand backend network.”

Data-centric Programming Helps ETH Zurich Team Win Gordon Bell Prize

Today ACM named a six-member team from ETH Zurich recipients of the 2019 ACM Gordon Bell Prize for their work on DaCe OMEN, a new framework for simulating the transport of electrical signals through nanoscale materials. “The ETH Zurich researchers simulated the 10,000-atom system 14 times faster than an earlier framework that was used for a 1,000- atom system. The DaCe OMEN code they developed for the simulation has been run on two top-6 hybrid supercomputers, reaching a sustained performance of 85.45 Pflop/s on 4,560 nodes of Summit (42.55% of the peak) in double precision, and 90.89 Pflop/s in mixed precision.”

CFD Leader NUMECA adopts Advania Data Centers for HPC-as-a-Service

Today Advania Data Centers (ADC) announced a strategic HPC service partnership with NUMECA International, a market leader in computational fluid dynamics CFD, multiphysics and optimization. “This HPCaaS partnership allows NUMECA to concentrate on its core software developments and support activities while ADC handles its HPC cluster management, power and cooling — ensuring peak performance at all times based on their Intel Xeon Scalable processor-based infrastructure.”

TYAN Launches AMD EPYC Server Platforms at SC19

Today TYAN rolled out the latest lineup of HPC and storage server platforms based on the AMD EPYC 7002 Series processors that are aimed at the datacenter market at SC19. “The 2nd Gen AMD EPYC processor was designed to provide customers with leadership in architecture, performance, and security,” said Scott Aylor, corporate vice president and general manager, Data Center Solutions Group, AMD. “We’re excited to see our partners, like TYAN, continue to build their portfolios around 2nd Gen EPYC to provide new capabilities for their customers and partners.”

2019 Demand for Rescale-managed Cloud HPC Exceeds All Previous Years Combined

Today Rescale announced  that Cloud High Performance Computing (HPC) has reached a major inflection point, with more server hours consumed this year on the Rescale platform than in all prior years combined in the company’s history. Every major cloud provider now offers integrations with the Rescale platform, including AWS, Microsoft Azure, IBM–as well as new offerings from Google and Oracle announced this week. New FedRAMP security and compliance milestones enhance the signal that mainstream companies can adopt cloud HPC.  

Dell Technologies taps AMD EPYC processors for Expanse Supercomputer at SDSC

Dell Technologies has been selected to power the next-generation supercomputer at the San Diego Supercomputer Center (SDSC), expected to deploy in mid-2020. “With the compute-dense PowerEdge C6525, including next-generation AMD EPYC processors NVIDIA GPUs, Expanse is projected to have a peak performance of up to five petaflops. This also nearly doubles the performance of SDSC’s current Comet system, allowing SDSC to support more researchers and projects.”

Mellanox Rocks the TOP500 with HDR InfiniBand at SC19

In this video, Gilad Shainer from Mellanox describes the company’s dominating  results on the TOP500 list of the world’s fastest supercomputers. “At SC19, Mellanox announced that 200 gigabit per second HDR InfiniBand accelerates 31% of the new 2019 InfiniBand systems on November’s TOP500 supercomputing list, demonstrating market demand for faster data speeds and smart interconnect technologies. Moreover, HDR InfiniBand connects the fastest TOP500 supercomputer built in 2019.”

Innovium High Performance Networking Solutions Move Big Data at SC19

Innovium is showcasing their TERALYNX-based high performance switches SC19 this week. These switches deliver the industry’s highest performance, lowest latency, and highest port radix with flexibility for 10G – 400G connectivity and breakthrough telemetry using Innovium’s programmable TERALYNX switch silicon.