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


How Zettar Transferred 1 Petabyte of Data in Just 34 Hours Using AIC Servers

In the world of HPC, moving data is a sin. That may be changing. “Just a few weeks ago, AIC announced the successful completion of a landmark, 1-petabyte transfer of data in 34 hours, during a recent test by Zettar that relied on the company’s SB122A-PH, 1U 10-bay NVMe storage server. The milestone was reached using a unique 5000-mile 100Gbps loop which is a SDN layer over a shared, production 100G network operated by the US DOE’s ESNet.”

AFRL Taps IBM to Build Brain-Inspired AI Supercomputer

Today IBM announced they are collaborating with the U.S. Air Force Research Laboratory (AFRL) on a first-of-a-kind brain-inspired supercomputing system powered by a 64-chip array of the IBM TrueNorth Neurosynaptic System. The scalable platform IBM is building for AFRL will feature an end-to-end software ecosystem designed to enable deep neural-network learning and information discovery.

Job of the Week: Computational Scientist at the University of Oregon

The University of Oregon High Performance Computing Research Core Facility (HPCRCF) is seeking a Computational Scientist in our Job of the Week. The HPCRCF is a new facility located on the campus of the UO in Eugene, Oregon. The mission of the HPCRCF is to support computational research at the UO and collaborating institutions, and is home to a new flagship research cluster.

One Stop Systems Showcases GPU and Flash Appliances at ISC

This week in Frankfurt, One Stop Systems showcased all-new HPC Appliances at ISC 2017. “One Stop Systems offers a wide variety of GPU and Flash appliances to support a range of customer needs,” Said Steve Cooper, CEO of One Stop Systems. “As GPU and flash technology continue to improve, OSS products are immediately able to accommodate the newest and most powerful GPUs and the highest capacity flash cards. Now we’re offering customers the opportunity to lease time on our HPC appliances through SkyScale. SkyScale utilizes OSS GPU appliances to build, configure, and manage dedicated systems strategically located in maximum-security facilities, allowing customers to focus on results while minimizing capital equipment investment.”

Podcast Brent Gorda on the State of Lustre at ISC 2017

In this RichReport podcast from the German Autobahn, Brent Gorda gives us a Lustre update from ISC 2017 in Frankfurt. “On the latest Top500 list, Lustre leads the pack with 9 systems out of the Top 10.”

Kim McMahon on her role as Diversity Chair at ISC 2018

In this video, Kim McMahon describes her new role as Diversity Chair for the ISC 2018 conference. “ISC has named me the Diversity Chair for ISC18. With this comes a position on the Steering Committee. While this is really cool in and of itself, I am very excited about the industry luminaries, experts, and innovators that I’ll get to meet and work with. The list of people on the ISC Steering Committee is a Who’s Who of the HPC industry! I am deeply honored with the trust ISC has placed in me and look forward to the upcoming year’s activities.”

HPC Vendors showcase plans for ARM products

Are you ready? ARM-based HPC systems will be available by the end of 2017. At ISC 2017, Penguin Computing and Bull both announced that they will deliver products tailored for the HPC industry. Both companies systems will feature Cavium’s ARMv8-based ThunderX2 platform.

NASA’s Develops SAFEGUARD Safety Net Technology for Drones

Flying drone aircraft are no longer rare sightings these day. In fact, their may be one above you somewhere taking pictures at this very moment. This begs the question: how do we keep our skies safe? “Scientists at NASA Langley Research Center have developed a breakthrough technology called Safeguard that can alleviate hazards with unmanned aircrafts (UA) flying beyond their authorized perimeters and into no-fly zones. Safeguard works by continuously detecting a UAs proximity to virtual perimeters established around no-fly-zones taking action to guarantee the perimeters are not breached.

Simply Beautiful: MareNorstum4 Supercomputer Sports 13.7 Petaflops

Over at Lenovo, Gavin O’Hara writes that the world’s most beautiful supercomputer center now sports a 13.7 Petaflop system so novel in design that it has captured the attention of the global HPC community. It landed at #13 on the TOP500 this week, and that’s just the beginning.

Deep Learning Frameworks Get a Performance Benefit from Intel MKL Matrix-Matrix Multiplication

Intel® Math Kernel Library 2017 (Intel® MKL 2017) includes new GEMM kernels that are optimized for various skewed matrix sizes. The new kernels take advantage of Intel® Advanced Vector Extensions 512 (Intel® AVX-512) and achieves high GEMM performance on multicore and many-core Intel® architectures, particularly for situations arising from deep neural networks..