MailChimp Developer

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


Hewlett Packard Enterprise Packs 8 GPUs into Apollo 6500 Server

In this video from the 2016 GPU Technology Conference, Greg Schmidt from Hewlett Packard Enterprise describes the new Apollo 6500 server. “With up to eight high performance NVIDIA GPU cards designed for maximum transfer bandwidth, the HPE Apollo 6500 System is purpose-built for deep learning applications. Its high ratio of GPUs to CPUs, dense 4U form factor and efficient design enable organizations to run deep learning recommendation algorithms faster and more efficiently, significantly reducing model training time and accelerating the delivery of real-time results, all while controlling costs.”

How HPE Makes GPUs Easier to Program for Data Scientists

In this video from the 2016 GPU Technology Conference, Rich Friedrich from Hewlett Packard Enterprise describes how the company makes it easier for Data Scientists to program GPUs. “In April, HPE announced a public, open-source version of the platform called the Cognitive Computing Toolkit. Instead of relying on the traditional CPUs that power most computers, the Toolkit runs on graphics processing units (GPUs), inexpensive chips designed for video game applications.”

Video: Lustre Community Release Update

Peter Jones from Intel presented this talk at LUG 2016 in Portland. “The OpenSFS Lustre Working Group (LWG) is the place the where the participants of OpenSFS come together to coordinate their software development efforts for the Lustre high-performance, Open Source, parallel filesystem. This includes planning and the roadmap for community releases of Lustre.”

Video: AMD ROC – Radeon Open Compute Platform

Gregory Stoner from AMD presented this talk at the HPC User Forum. “With the announcement of the Boltzmann Initiative and the recent releases of ROCK and ROCR, AMD has ushered in a new era of Heterogeneous Computing. The Boltzmann initiative exposes cutting edge compute capabilities and features on targeted AMD/ATI Radeon discrete GPUs through an open source software stack. The Boltzmann stack is comprised of several components based on open standards, but extended so important hardware capabilities are not hidden by the implementation.”

Video: Europe’s Fastest Supercomputer and the World Around It

Michael Resch from HLRS gave this rousing talk at the HPC User Forum. “HLRS supports national and European researchers from science and industry by providing high-performance computing platforms and technologies, services and support. Supercomputer Hazel Hen, a Cray XC40-system, is at the heart of the HPC system infrastructure of the HLRS. With a peak performance of 7.42 Petaflops (quadrillion floating point operations per second), Hazel Hen is one of the most powerful HPC systems in the world (position 8 of TOP500, 11/2015) and is the fastest supercomputer in the European Union. The HLRS supercomputer, which was taken into operation in October 2015, is based on the Intel Haswell Processor and the Cray Aries network and is designed for sustained application performance and high scalability.”

Slidecast: Advantages of Offloading Architectures for HPC

In this slidecast, Gilad Shainer from Mellanox describes the advantages of InfiniBand and the company’s off-loading network architecture for HPC. “The path to Exascale computing is clearly paved with Co-Design architecture. By using a Co-Design approach, the network infrastructure becomes more intelligent, which reduces the overhead on the CPU and streamlines the process of passing data throughout the network. A smart network is the only way that HPC data centers can deal with the massive demands to scale, to deliver constant performance improvements, and to handle exponential data growth.”

ORiGAMI – Oak Ridge Graph Analytics for Medical Innovation

Rangan Sukumar from ORNL presented this talk at the HPC User Forum in Tucson. “ORiGAMI is a tool for discovering and evaluating potentially interesting associations and creating novel hypothesis in medicine. ORiGAMI will help you “connect the dots” across 70 million knowledge nuggets published in 23 million papers in the medical literature. The tool works on a ‘Knowledge Graph’ derived from SEMANTIC MEDLINE published by the National Library of Medicine integrated with scalable software that enables term-based, path-based, meta-pattern and analogy-based reasoning principles.”

Jetstream – Adding Cloud-based Computing to the National Cyberinfrastructure

Matt Vaughan from TACC presented this talk at the HPC User Forum. “Jetstream is the first user-friendly, scalable cloud environment for XSEDE. The system enables researchers working at the “long tail of science” and the creation of truly customized virtual machines and computing architectures. It has a web-based user interface integrated with XSEDE via Globus Auth. The architecture is derived from the team’s collective experience with CyVerse Atmosphere, Chameleon and Quarry. The system also fosters reproducible, sharable computing with geographically isolated clouds located at Indiana University and TACC.”

Video: How to Build a Neural Net in 4 Minutes

In this video, Siraj Rival from Twilio presents a quick tutorial on How to Build a Neural Net in 4 Minutes. Siraj describes himself as the Bill Nye of Computer Science.

Using High Performance Interconnects in Dynamic Environments

“Over the last years the OFA community has shown the potential of using high performance networks (InfiniBand) to boost the performance of virtualized cloud environments, however, the network reconfiguration challenges still continue to exist. In this session we present the work we have been doing on InfiniBand subnet management and routing, in the context of dynamic cloud environments. This work includes, but not limited to, techniques in order to provide better management scalability when virtual machines are live migrating, tenant network isolation in multi-tenant environments, and fast performance-driven network reconfiguration.”