High-Performance Computing News Analysis | insideHPC

Humans are very good at visual pattern recognition especially when it comes to facial features and graphic symbols and identifying a specific person or associating a specific symbol with an associated meaning. It is in these kinds of scenarios where deep learning systems excel. Clearly identifying each new person or symbol is more efficiently achieved by a training methodology than by needing to reprogram a conventional computer or explicitly update database entries.The post Examples of Deep Learning Industrialization appeared first on insideHPC.

"Our collaborative role in these exascale applications projects stems from our laboratory’s long-term strategy in co-design and our appreciation of the vital role of high-performance computing to address national security challenges,” said John Sarrao, associate director for Theory, Simulation and Computation at Los Alamos National Laboratory. “The opportunity to take on these scientific explorations will be especially rewarding because of the strategic partnerships with our sister laboratories.”The post Funding Boosts Exascale Research at LANL appeared first on insideHPC.

In this podcast, the Radio Free HPC team discusses the recent news that Intel has sold its controlling stake in McAfee and that NSF has funded the next generation of XSEDE.The post Radio Free Looks at Exascale Application Challenges in the Wake of XSEDE 2.0 Funding appeared first on insideHPC.

A consortium of European researchers and technology companies recently completed the EU-funded SAVE project, aimed at simplifying the execution data-intensive applications on complex hardware architectures. Funded by the European Commission’s Seventh Framework Programme (FP7), the project was launched in 2013, under the project name ‘Self-Adaptive Virtualization-Aware High-Performance/Low-Energy Heterogeneous System Architectures’ (SAVE). The project, which was completed at the start of this month, has led to innovations in hardware, software and operating system (OS) components.The post European SAVE Project Streamlines Data Intensive Computing appeared first on insideHPC.

Leonardo Flores from the European Commission presented this talk at the HPC User Forum. "The Cloud Initiative will make it easier for researchers, businesses and public services to fully exploit the benefits of Big Data by making it possible to move, share and re-use data seamlessly across global markets and borders, and among institutions and research disciplines. Making research data openly available can help boost Europe's competitiveness, especially for start-ups, SMEs and companies who can use data as a basis for R&D and innovation, and can even spur new industries."The post EU HPC Strategy and the European Cloud Initiative appeared first on insideHPC.

Yutaka Ishikawa from Riken AICS presented this talk at the HPC User Forum. "Slated for delivery sometime around 2022, the ARM-based Post-K Computer has a performance target of being 100 times faster than the original K computer within a power envelope that will only be 3-4 times that of its predecessor. RIKEN AICS has been appointed as the main organization for leading the development of the Post-K."The post Video: Japan’s Post K Computer appeared first on insideHPC.

Coming to SC16? Intel is hosting a one-day Lustre training session on Friday November 18th, 2016 at the Sheraton in Salt Lake City. "You can expect the top Lustre experts to spend the day with on the training topics that you care about."The post Lustre Training Day Coming to SC16 in Salt Lake City appeared first on insideHPC.

Supermicro’s density optimized 4U SuperServer 4028GR-TR(T)2 supports up to 10 PCI-E Tesla P100 accelerators for up to 210 TFLOPS FP16 peak performance with GPU Direct RDMA support. Supermicro’s innovative and GPU optimized single root complex PCI-E design is proven to dramatically improve GPU peer-to-peer communication efficiency over QPI and PCI-E links, with up to 21% higher QPI throughput and 60% lower latency compared to previous generation products. These 4U SuperServers support dual Intel Xeon processor E5-2600 v4/v3 product families, up to 3TB DDR4-2400MHz memory, optional dual onboard 10GBase-T ports, and redundant Titanium Level (96%) digital power supplies.The post Supermicro Shipping Servers with NVIDIA Tesla P100 GPUs appeared first on insideHPC.

Today Juniper Networks announced it has been selected by the National Center for Atmospheric Research (NCAR) to provide the networking infrastructure for a new supercomputer that will be used by researchers to predict climate patterns and assess the effects of global warming.The post Juniper Networks Powers Climate Research at NCAR appeared first on insideHPC.

"Nimbis was founded in 2008 by HPC industry veterans Robert Graybill and Brian Schott to act as the first nationwide brokerage clearinghouse for a broad spectrum of integrated cloud-based HPC platforms and applications. Our fully integrated online Technical Computing Marketplace comprises several stores hosting modeling and simulation applications on HPC platforms in the cloud."The post Video: HPC in the Cloud appeared first on insideHPC.

"With the Yottabyte Research Cloud, researchers will be able to ask more questions, faster, of the ever-expanding and massive sets of data collected for their work," said Yottabyte CEO Paul Hodges. "We are very pleased to be a part of the diverse and challenging research environment at U-M. This partnership is a great opportunity to develop and refine computing tools that will increase the productivity of U-M's world class researchers."The post Yottabyte Powers Data-Intensive Research at University of Michigan appeared first on insideHPC.

Today One Stop Systems (OSS) announced that its High Density Compute Accelerator (HDCA) and its Express Box 3600 (EB3600) are now available for purchase with the NVIDIA Tesla P100 for PCIe GPU. These high-density platforms deliver teraflop performance with greatly reduced cost and space requirements. The HDCA supports up to 16 Tesla P100s and the EB3600 supports up to 9 Tesla P100s. The Tesla P100 provides 4.7 TeraFLOPS of double-precision performance, 9.3 TeraFLOPS of single-precision performance and 18.7 TeraFLOPS of half-precision performance with NVIDIA GPU BOOST technology.The post One Stop Systems Shipping Platforms with NVIDIA Tesla P100 for PCIe appeared first on insideHPC.

Today Fujitsu announced that the company has received an order for an experiment-analysis system from Kamioka Observatory, part of the Institute for Cosmic Ray Research (ICRR) at the University of Tokyo. The system is destined for Kamioka Observatory's Super-Kamiokande facility, which is helping to shed light on the workings of the universe through the observation of neutrinos, and is scheduled to go operational in March 2017.The post Fujitsu Cluster to Power Super-Kamiokande Neutrino Experiments appeared first on insideHPC.

"An environment that assists in deep learning usually consists of algorithms that can draw conclusions from data that is run at very high speeds. Processors such as the Intel Xeon Phi Processor that contain a significant number of processing cores and operate in a SIMD mode are critical to these new environments. With the Intel Xeon Phi processor, new insights can be discovered from either existing data or new data sources."The post Deep Learning with the Intel Xeon Phi Processor appeared first on insideHPC.

Today Argonne announced that the Lab is leading a pair of newly funded applications projects for the Exascale Computing Project (ECP). The announcement comes on the heels of news that ECP has funded a total of 15 application development proposals for full funding and seven proposals for seed funding, representing teams from 45 research and academic organizations.The post Argonne to Develop Applications for ECP Exascale Computing Project appeared first on insideHPC.

Ed Turkel and Percy Tzelnic from Dell Technologies presented this pair of talks at the HPC User Forum in Austin. "This week, Dell Technologies announced completion of the acquisition of EMC Corporation, creating a unique family of businesses that provides the essential infrastructure for organizations to build their digital future, transform IT and protect their most important asset, information. This combination creates a $74 billion market leader with an expansive technology portfolio."The post Dell & EMC in HPC – The Journey so far and the Road Ahead appeared first on insideHPC.

“The University’s researchers are making landmark discoveries in fields spanning human heritable disease, cancer, agriculture and biofuels manufacture – and they depend on our IT team to provide them with the fastest, most efficient data storage and compute systems to support their data-heavy work,” said Professor David Abramson, University of Queensland Research Computing Center director. “Our IBM, SGI (DMF) and DDN-based data fabric allows us to deliver ultra-fast multi-site data access without requiring any extra intervention from researchers and helps us to ensure our scientists can focus their time on potentially life-saving discoveries.”The post DDN Powers High Performance Data Storage Fabric at University of Queensland appeared first on insideHPC.

Today IBM unveiled a series of new servers designed to help propel cognitive workloads and to drive greater data center efficiency. Featuring a new chip, the Linux-based lineup incorporates innovations from the OpenPOWER community that deliver higher levels of performance and greater computing efficiency than available on any x86-based server. "Collaboratively developed with some of the world’s leading technology companies, the new Power Systems are uniquely designed to propel artificial intelligence, deep learning, high performance data analytics and other compute-heavy workloads, which can help businesses and cloud service providers save money on data center costs."The post New OpenPOWER Servers Accelerate Deep Learning with NVLink appeared first on insideHPC.

Paul Messina presented this talk at the HPC User Forum in Austin. "The Exascale Computing Project (ECP) is a collaborative effort of the Office of Science (DOE-SC) and the National Nuclear Security Administration (NNSA). As part of President Obama’s National Strategic Computing initiative, ECP was established to develop a new class of high-performance computing systems whose power will be a thousand times more powerful than today’s petaflop machines."The post Video: The ECP Exascale Computing Project appeared first on insideHPC.

Today Dell Technologies announced completion of the acquisition of EMC Corporation, creating a unique family of businesses that provides the essential infrastructure for organizations to build their digital future, transform IT and protect their most important asset, information. This combination creates a $74 billion[i] market leader with an expansive technology portfolio that solves complex problems for customers in the industry’s fast-growing areas of hybrid cloud, software-defined data center, converged infrastructure, platform-as-a-service, data analytics, mobility and cybersecurity.The post It’s “Day 1” for Dell Technologies with New Branding appeared first on insideHPC.

Achieving better scalability and performance at Exascale will require full data reach. Without this capability, onload architectures force all data to move to the CPU before allowing any analysis. The ability to analyze data everywhere means that every active component in the cluster will contribute to the computing capabilities and boost performance. In effect, the interconnect will become its own “CPU” and provide in-network computing capabilities.The post Network Co-design as a Gateway to Exascale appeared first on insideHPC.

Today Lawrence Berkeley National Laboratory announced that LBNL scientists will lead or play key roles in developing 11 critical research applications for next-generation supercomputers as part of DOE's Exascale Computing Project (ECP).The post Berkeley Lab to Develop Key Applications for ECP Exascale Computing Project appeared first on insideHPC.

Earl Joseph presented this talk at the HPC User Forum in Austin. "HPC is still expected to be a strong growth market going forward. IDC is forecasting a 7.7 percent growth from 2015 to 2019. We're projecting the 2019 HPC Market will exceed $15 Billion."The post Video: IDC HPC Market Update appeared first on insideHPC.

"Aligning ourselves with an organization like Microsoft further supports our mission of enabling companies to manage their application workloads whether on-premise or in a hybrid or cloud context like Microsoft Azure,” said Gary Tyreman, CEO, Univa. “Univa is working hard to bridge the gap from traditional IT environments to the new hybrid world by helping enterprises reduce complexity and run workloads more efficiently in the cloud.”The post Univa Joins up with Microsoft Enterprise Cloud Alliance appeared first on insideHPC.

"These application development awards are a major first step toward achieving mission critical application readiness on the path to exascale,” said ECP director Paul Messina. “A key element of the ECP’s mission is to deliver breakthrough HPC modeling and simulation solutions that confidently deliver insight and predict answers to the most critical U.S. problems and challenges in scientific discovery, energy assurance, economic competitiveness, and national security,” Messina said. “Application readiness is a strategic aspect of our project and foundational to the development of holistic, capable exascale computing environments.”The post Exascale Computing Project (ECP) Awards $39.8 million for Application Development appeared first on insideHPC.

Hailing from Norway, big-memory appliance maker Numascale has been a fixture at the ISC conference since the company’s formation in 2008. At ISC 2016, Numascale was noticeably absent from the show and the word on the street was that the company was retooling their NumaConnect™ technology around NVMe. To learn more, we caught up with Einar Rustad, Numascale’s CTO.The post Interview: Numascale to Partner with OEMs on Big Memory Server Technology appeared first on insideHPC.

Submissions are now open for the ISC 2017 conference Research Paper Sessions, which aim to provide first-class opportunities for engineers and scientists in academia, industry and government to present research that will shape the future of high performance computing. Submissions will be accepted through December 2, 2016.The post Call for Research Papers: ISC 2017 appeared first on insideHPC.

Today Movidius announced that the company is being acquired by Intel. As a deep learning startup, Movidius's mission is to give the power of sight to machines.The post Intel Acquires Movidius to Bolster AI appeared first on insideHPC.

With Intel Scalable System Framework Architecture Specification and Reference Designs, the company is making it easier to accelerate the time to discovery through high-performance computing. The Reference Architectures (RAs) and Reference Designs take Intel Scalable System Framework to the next step—deploying it in ways that will allow users to confidently run their workloads and allow system builders to innovate and differentiate designsThe post Facilitate HPC Deployments with Reference Designs for Intel Scalable System Framework appeared first on insideHPC.

"This release will allow aerospace stress analysts to do their tasks in a much more efficient manner. We really focused on understanding the desired workflows and creating an environment to easily move between CAD models, CAE models and results, and external tools such as Microsoft Excel,” said Dr. Robert Yancey, Altair Vice President of Aerospace. “We look forward to working with our Aerospace customers to help them implement their workflows in the streamlined HyperWorks environment.”The post Altair Updates HyperWorks for Aerospace appeared first on insideHPC.

Charles W. Nakhleh from LANL presented this talk at the 2016 DOE NNSA SSGF Annual Program Review. "This talk will explore some of the future opportunities and exciting scientific and technological challenges in the National Nuclear Security Administration Stockpile Stewardship Program. The program’s objective is to ensure that the nation's nuclear deterrent remains safe, secure and effective. Meeting that objective requires sustained excellence in a variety of scientific and engineering disciplines and has led to remarkable advances in theory, experiment and simulation."The post The Challenges and Rewards of Stockpile Stewardship appeared first on insideHPC.

Applications are now open for the annual SuperComputing Camp in Colombia. The five-day camp takes place Oct. 16-21 at CIBioFI at Universidad del Valle in Santiago de Cali.The post SuperComputing Camp Returns to Colombia appeared first on insideHPC.

The Piz Daint supercomputer at the Swiss National Supercomputing Centre (CSCS) is again assisting researchers in competition for the prestigious Gordon Bell prize. "Researchers led by Peter Vincent from Imperial College London have made this year’s list of finalists for the Gordon Bell prize, with the backing of Piz Daint at the Swiss National Supercomputing Centre. The prize is awarded annually in November at SC, the world’s largest conference on supercomputing. It honors the success of scientists who are able to achieve very high efficiencies for their research codes running on the fastest supercomputer architectures currently available."The post Powering Aircraft CFD with the Piz Daint Supercomputer appeared first on insideHPC.

Rick Wagner from SDSC presented this talk at the the 4th Annual MVAPICH User Group. "At SDSC, we have created a novel framework and infrastructure by providing virtual HPC clusters to projects using the NSF sponsored Comet supercomputer. Managing virtual clusters on Comet is similar to managing a bare-metal cluster in terms of processes and tools that are employed. This is beneficial because such processes and tools are familiar to cluster administrators."The post Video: User Managed Virtual Clusters in Comet appeared first on insideHPC.

Over at NASA, Michelle Moyers writes that the 2016 NASA Software of the Year Award has gone to Pegasus 5, a revolutionary CFD tool. "Developed in-house by a team led by aerospace engineer Stuart Rogers from NASA Ames, Pegasus 5 has been used for aerodynamic modeling and simulation by nearly every NASA program over the past 15 years, including the space shuttle, the next-generation Orion spacecraft and Space Launch System, and commercial crew programs."The post Pegasus 5 CFD Program Wins NASA’s Software of the Year appeared first on insideHPC.

In this podcast, the Radio Free HPC team previews the HPC User Forum & StartupHPC Events coming up in the Fall of 2016.The post Radio Free HPC Previews the HPC User Forum & StartupHPC Events appeared first on insideHPC.

Today ALA Services announced it has acquired Adaptive Computing of Provo, Utah. "Adaptive Computing is adding proven growth expertise and infrastructure through this acquisition by ALA Services," said Marty Smuin, CEO of Adaptive Computing. "Arthur L. Allen brings deep insights and a proven process he has used to drive success. We look forward to accelerating our business and improving value to our customers by combining ALA's expertise with Adaptive Computing's leading technology and great position within the market."The post ALA Services Acquires Adaptive Computing appeared first on insideHPC.

"By modeling the power system in depth and detail, NREL has helped reset the conversation about how far we can go operationally with wind and solar in one of the largest power systems in the world," said the Energy Department's Charlton Clark, a DOE program manager for the study. "Releasing the production cost model, underlying data, and visualization tools alongside the final report reflects our commitment to giving power system planners, operators, regulators, and others the tools to anticipate and plan for operational and other important changes that may be needed in some cleaner energy futures."The post Supercomputing Alternative Energy in the Eastern Power Grid appeared first on insideHPC.

Miles Lubin from presented this talk at the CSGF Annual Program Review. "JuMP is an open-source software package in Julia for modeling optimization problems. In less than three years since its release, JuMP has received more than 50 citations and has been used in at least 10 universities for teaching. We tell the story of how JuMP was developed, explain the role of the DOE CSGF and high-performance computing, and discuss ongoing extensions to JuMP developed in collaboration with DOE labs."The post Video: JuMP – A Modeling Language for Mathematical Optimization appeared first on insideHPC.

Scientists at the Energy Department's National Renewable Energy Laboratory (NREL) discovered a use for perovskites that could propel the development of quantum computing. "Considerable research at NREL and elsewhere has been conducted into the use of organic-inorganic hybrid perovskites as a solar cell. Perovskite systems have been shown to be highly efficient at converting sunlight to electricity. Experimenting on a lead-halide perovskite, NREL researchers found evidence the material could have great potential for optoelectronic applications beyond photovoltaics, including in the field of quantum computers."The post NREL Discovery Could Propel Quantum Computing appeared first on insideHPC.

Today, the National Geospatial-Intelligence Agency and NSF released 3-D topographic maps that show Alaska’s terrain in greater detail than ever before. Powered by the Blue Waters supercomputer, the maps are the result of a White House Arctic initiative to inform better decision-making in the Arctic. "We can’t live without Blue Waters now,” said Paul Morin, head of the University of Minnesota’s Polar Geospatial Center. “The supercomputer itself, the tools the Blue Waters team at NCSA developed, the techniques they’ve come up with in using this hardware. Blue Waters is changing the way digital terrain is made and that is changing how science is done in the Arctic.”The post Supercomputing 3D Elevation Maps of Alaska on Blue Waters appeared first on insideHPC.

Engineers at Sandia are developing new datacenter cooling technologies that could save millions of gallons of water nationwide.The post Saving Water with Sandia’s New Datacenter Cooling Technology appeared first on insideHPC.

James Reinders presented this talk at the 2016 Argonne Training Program on Extreme-Scale Computing. Reinders is the author of multiple books on parallel programming. His most recent book, entitled Intel Xeon Phi Processor High Performance Programming: Knights Landing Edition 2nd Edition, was co-authored by James Jeffers and Avinash Sodani.The post Video: SIMD, Vectorization, and Performance Tuning appeared first on insideHPC.

EPSRC and Cray have signed an agreement to add a Cray XC40 Development System with Intel Xeon Phi processors to ARCHER, the UK National Supercomputing Service. "The new Development system will have a very similar environment to the main ARCHER system, including Cray's Aries interconnect, operating system and Cray tools, meaning that interested users will enjoy a straightforward transition."The post Cray to Add Intel Xeon Phi to Archer Supercomputing Service in the UK appeared first on insideHPC.

"We have enhanced Bright Cluster Manager 7.3 so our customers can quickly and easily deploy new deep learning techniques to create predictive applications for fraud detection, demand forecasting, click prediction, and other data-intensive analyses,” said Martijn de Vries, Chief Technology Officer of Bright Computing. “Going forward, customers using Bright to deploy and manage clusters for deep learning will not have to worry about finding, configuring, and deploying all of the dependent software components needed to run deep learning libraries and frameworks.”The post New Bright for Deep Learning Solution Designed for Business appeared first on insideHPC.

"In the HPC domain, Python can be used to develop a wide range of applications. While tight loops may still need to be coded in C or FORTRAN, Python can still be used. As more systems become available with coprocessors or accelerators, Python can be used to offload the main CPU and take advantage of the coprocessor. pyMIC is a Python Offload Module for the Intel Xeon Phi Coprocessor and is available at popular open source code repositories."The post Python and HPC appeared first on insideHPC.

Today Dell Inc. and EMC Corp. announced that they intend to close the transaction to combine Dell and EMC on Wednesday, September 7, 2016. The name of the newly combined company will be Dell Technologies.The post Dell and EMC Transaction to Close on September 7, 2016 appeared first on insideHPC.

"Bridges has enabled early scientific successes, for example in metagenomics, organic semiconductor electrochemistry, genome assembly in endangered species, and public health decision-making. Over 2,300 users currently have access to Bridges for an extremely wide range of research spanning neuroscience, machine learning, biology, the social sciences, computer science, engineering, and many other fields."The post Bridges Supercomputer Enters Production at PSC appeared first on insideHPC.

The move to network offloading is the first step in co-designed systems. A large amount of overhead is required to service the huge number of packets required for modern data rates. This amount of overhead can significantly reduce network performance. Offloading network processing to the network interface card helped solve this bottleneck as well as some others.The post Co-Design Offloading appeared first on insideHPC.

Paul Messina presented this talk at the 2016 Argonne Training Program on Extreme-Scale Computing. "The President's NSCI initiative calls for the development of Exascale computing capabilities. The U.S. Department of Energy has been charged with carrying out that role in an initiative called the Exascale Computing Project (ECP). Messina has been tapped to lead the project, heading a team with representation from the six major participating DOE national laboratories: Argonne, Los Alamos, Lawrence Berkeley, Lawrence Livermore, Oak Ridge and Sandia. The project program office is located at Oak Ridge.The post Paul Messina Presents: A Path to Capable Exascale Computing appeared first on insideHPC.