HPC Wire

Subscribe to HPC Wire feed
Since 1987 - Covering the Fastest Computers in the World and the People Who Run Them
Updated: 6 hours 9 min ago

NVM Express, Inc. to Host NVMe over Fabrics Webcast on Sept. 20

Wed, 09/20/2017 - 10:13

WAKEFIELD, Mass., Sept. 20, 2017 — NVM Express, Inc., the organization that developed the NVM Express (NVMe) and NVMe Management Interface (NVMe-MI) specifications for accessing solid-state drives (SSDs) on a PCI Express (PCIe) bus as well over Fabrics, will lead an educational webcast titled “NVMe over Fabrics: Market Uptake, Benefits and Use Cases” on Wednesday, Sept. 20, 2017, at noon Eastern. There is no charge to attend, but advance registration is required.

WHAT: The webcast will cover how NVMe over Fabrics (NVMe-oF™) takes the NVMe beyond the single-computer paradigm to be useful in the cloud, clusters, distributed systems, and megawebsites. The presentation will examine how the NVMe-oF specification can unleash the benefits of NVMe drives in a scalable manner by leveraging the capabilities of mainstream high-performance interconnects.

Attendees will learn about the benefits of using different fabrics as the platform, such as Fibre Channel, RoCE, iWarp, and Infiniband. The presentation will explore the next frontier of the NVMe-oF specification, including use at the TCP layer or deployed as NVMe SSDs in the data center. Furthermore, the webcast discussion will advance beyond transports to discuss new storage architectures and use cases.

WHO: Presenter is NVM Express’ Brandon Hoff, distinguished software architect, Broadcom

WHERE: https://www.brighttalk.com/webcast/12367/275423

WHEN: Wednesday, Sept. 20, 2017, at noon Eastern

About NVM Express, Inc.

With more than 100 members, NVM Express, Inc. is a non-profit organization focused on enabling broad ecosystem adoption of high performance and low latency non-volatile memory (NVM) storage through a standards-based approach. The organization offers an open collection of NVM Express (NVMe™) specifications and information to fully expose the benefits of non-volatile memory in all types of computing environments from mobile to data center. NVMe-based specifications are designed from the ground up to deliver high bandwidth and low latency storage access for current and future NVM technologies. For more information, visit http://www.nvmexpress.org.

Source: NVM Express, Inc.

The post NVM Express, Inc. to Host NVMe over Fabrics Webcast on Sept. 20 appeared first on HPCwire.

DeepL Anchors Neural Machine Translator at Verne Global’s HPC-Optimised Data Center

Wed, 09/20/2017 - 10:05

LONDON & KEFLAVIK, Iceland, Sept. 20, 2017 — Verne Global, a provider of highly optimised, secure, and 100% renewably powered data center solutions, today announced that DeepL has deployed its 5.1 petaFLOPS supercomputer in its campus. Designed to support DeepL’s artificial intelligence (AI) driven, neural network translation service, this supercomputer is viewed by many as the world’s most accurate and natural-sounding machine translation service. Verne Global was selected because of the following factors:

  • The innovative campus design specialised to support HPC and other intensive compute environments driven by the rise in AI, machine learning and big data analytics
  • The expertise and technical knowledge of the Verne Global team, and
  • Verne Global’s access to Iceland’s abundant, renewable power and its highly reliable, low-cost energy grid.

“For DeepL, we needed a data center optimised for high-performance computing (HPC) environments and determined that our needs could not be met in Germany. Verne Global’s Icelandic campus provides us with the scalability, flexibility and technical resources we need. In addition, the abundance of low-cost renewable energy and free cooling will allow us to train DeepL’s neural networks at lower cost and faster scalability,” says Jaroslaw Kutylowski, CTO of DeepL. “Verne Global’s team has a high level of technical expertise, which helps us to implement ad hoc requests quickly and easily. I’ve never seen such an excellent cooperation before.”

On the supercomputer located within Verne Global’s campus, DeepL trains the neuronal translation networks based on collected data sets. As DeepL learns, the network leverages AI to examine millions of translations and learn independently how to translate with the right grammar and structure.

“We are pleased that our HPC-optimised campus was the ideal location for DeepL’s supercomputer. Our location in Iceland provides a low and stable energy price with the highest possible availability and scalability – criteria that are indispensable for computational and power-intensive applications,” says Tate Cantrell, Chief Technology Officer of Verne Global. “We are seeing growing interest from companies using AI tools, such as deep neural network (DNN) applications, to revolutionise how they move their businesses forward, create change, and elevate how we work, live and communicate.”

The market for AI, machine learning and cognitive computing is expanding rapidly. According to a recent paper, “Artificial Intelligence, The Next Digital Frontier?”, issued by the McKinsey Global Institute, the total annual external investment in AI was between $8B to $12B in 2016, with machine learning attracting nearly 60% of that investment. McKinsey also states that, “A confluence of developments is driving this new wave of AI development. Computer power is growing, algorithms and AI models are becoming more sophisticated, and, perhaps most important of all, the world is generating once unimaginable volumes of the fuel that powers AI—data. Billions of gigabytes every day, collected by networked devices ranging from web browsers to turbine sensors.”

Verne Global’s data center, located on a former NATO base in Iceland, draws its electricity from hydroelectric and geothermal energy. The cool, temperate climate in Iceland enables free cooling, that when combined with the low-cost, renewable power, means that companies can save more than 70% on the total cost of operations for their compute resources over less optimal locations within the US, UK and continental Europe. The combination of innovative, technical design and an optimal location make Verne Global one of the worlds most efficient data center campuses.

Source: Verne Global

The post DeepL Anchors Neural Machine Translator at Verne Global’s HPC-Optimised Data Center appeared first on HPCwire.

DataSite Deploys High Performance Computing Solution for Enterprise Customer in Orlando

Wed, 09/20/2017 - 09:38

ORLANDO, Fla., Sept. 20, 2017 — DataSite, a wholesale data center space and network service provider that owns and operates world-class colocation facilities in Atlanta, Boise and Orlando, announces it has deployed a high-performance compute (HPC) system, for a major hospitality corporation, from its Orlando data center. The mission critical environment will accommodate air cooled power densities of 20kW+ per rack.

Market research firm, IDC, projects the HPC market will grow from $23.1B in 2016 to $31.4B in 2019, fueled by the desire to reap the of benefits of power hungry servers, storage, middleware, and applications. DataSite has responded to soaring demand for HPC with a multitude of solutions to accommodate high density workloads across its three data centers. Chimney cabinets, which direct hot air up and into the active containment plenum; extra rack depth to accommodate hot air containment; and the use of blanking panels and materials are all part of this sophisticated, integrated solution.

“DataSite is thrilled to grow its deployment HPC solutions to a growing list of customers,” comments Rob Wilson, Executive Vice President for DataSite. “We are excited to support other industries as they embark on HPC initiatives. DataSite is well-equipped to sensitively meet all custom HPC requirements that call for up to 90kW per rack armed with expertise in thermodynamics and a wide arsenal of technological solutions at its disposal.”

Visit www.datasitecolo.com to learn more about the company’s data centers and HPC initiatives.

About DataSite

DataSite offers secure world-class facilities in Atlanta, Boise and Orlando, capable of accommodating varying needs in wholesale data center space. DataSite data centers are a unique blend of purpose-built, specially constructed data center facilities and expertly managed data center infrastructure designed to offer affordable colocation options that meet the demanding power density and up-time requirements of the modern computing environment. DataSite’s Tier III data center design provides completely redundant and continually operating facilities that are concurrently maintainable with zero scheduled downtime.

Source: DataSite

The post DataSite Deploys High Performance Computing Solution for Enterprise Customer in Orlando appeared first on HPCwire.

Altair Extends Strategic Relationship with HPE

Wed, 09/20/2017 - 09:27

TROY, Mich., Sept. 19, 2017 — Altair announces today that it has entered into a multi-year original equipment manufacturing (OEM) agreement with HPE. This agreement represents an expansion of the long-term partnership between HPE and SGI (whom HPE recently acquired). HPE will now be able to include Altair’s PBS Professional workload manager and job scheduler on all of HPE’s high performance computing (HPC) systems, ensuring scalability of price and performance as system sizes and CPU-core counts continue to increase.

“We are delighted to strengthen our strategic collaboration with HPE,” said Sam Mahalingam, Chief Technical Officer for Enterprise Solutions at Altair. “With PBS Professional as its premier workload management software supplier, HPE will be able to provide our common customers with a powerful solution to meet their growing HPC requirements.”

PBS Professional gives HPE cluster users a more efficient, reliable solution for HPC workload management. As an HPE-integrated product, PBS Professional optimizes job scheduling on HPE Apollo and HPE SGI servers to achieve the highest levels of system utilization. PBS Professional is also integrated with HPE’s HPC system management solutions: HPE Insight Cluster Management Utility for (CMU) for HPE Apollo and HPE ProLiant platforms as well as HPE SGI Management Suite for HPE SGI 8600 systems.

“Altair’s PBS Professional is an established leader in HPC workload management,” said Bill Mannel, Vice President and General Manager for HPC and AI segment solutions at HPE. “We look forward to leveraging this agreement to give our customers access to an attractive PBS Professional offering to manage job scheduling and maximize system utilization on HPE’s industry leading HPC infrastructure.”

As the hardware vendor with the largest HPC market share, HPE offers the broadest spectrum of high-performance computing solutions, from workgroup and departmental servers to systems designed for the engineering enterprise and supercomputing centers (for more information please visit www.hpe.com/info/hpc).

Altair has served the HPC market for decades with award-winning workload management, engineering, and cloud computing software. Used by thousands of companies worldwide, PBS Professional enables engineers in HPC environments to improve productivity, optimize resource utilization and efficiency, and simplify the process of cluster workload management.

Click here for more information about HPE and Altair collaborations.

Customers can already obtain PBS Professional through HPE and its authorized resellers under the terms of the OEM agreement.

About Altair

Altair is focused on the development and broad application of simulation technology to synthesize and optimize designs, processes and decisions for improved business performance. Privately held with more than 2,600 employees, Altair is headquartered in Troy, Michigan, USA and operates 68 offices throughout 24 countries. Today, Altair serves more than 5,000 corporate clients across broad industry segments.

Source: Altair

The post Altair Extends Strategic Relationship with HPE appeared first on HPCwire.

HSA and ROCm Architectures to be Highlighted at Next Week’s CppCon

Wed, 09/20/2017 - 09:19

BEAVERTON, Ore., Sept. 20, 2017 — The HSA (Heterogeneous System Architecture) Foundation and Foundation member AMD will be providing a comprehensive session on HSA technologies and AMD’s ROCm architecture at next week’s CppCon. The conference will be held from Sept. 24-29 in Bellevue, WA at the Meydenbauer Conference Center.

CppCon is an annual gathering for the worldwide C++ community and is geared to appeal to anyone from C++ novices to experts.

The presentation by AMD Fellow Paul Blinzer is included as part of a session on ‘concurrency and parallelism’ running from 8:30-10 PM on Tuesday, Sept. 28 at the Meydenbauer Conference Center, Harvard, Room #406. Attendees will learn about what allows these architectures to use computational hardware accelerators like GPUs, DSPs and others with native C++, without resorting to proprietary APIs, programming libraries or limited language features.

Heterogeneous System Architecture (HSA) is a standardized platform design that unlocks the performance and power efficiency of the parallel computing engines found in most modern electronic devices. It provides an ideal mainstream platform for next-generation SoCs in a range of applications including artificial intelligence.

For more information on the presentation and to register, please see https://cppcon.org/registration/.

For more information, including a full list of speakers, supporting organizations and sponsors please visit: https://cppcon.org/cppcon-2017-program/

About Paul Blinzer

Paul Blinzer works on a wide variety of Platform System Software architecture projects and specifically on the Heterogeneous System Architecture (HSA) System Software at Advanced Micro Devices, Inc. (AMD) as a Fellow in the System Software group. Living in the Seattle, WA area, during his career he has worked in various roles on system level driver development, system software development, graphics architecture, graphics & compute acceleration since the early ’90s. Paul is the chairperson of the “System Architecture Workgroup” of the HSA Foundation. He has a degree in Electrical Engineering (Dipl.-Ing) from TU Braunschweig, Germany.

About the HSA Foundation

The HSA (Heterogeneous System Architecture) Foundation is a non-profit consortium of SoC IP vendors, OEMs, Academia, SoC vendors, OSVs and ISVs, whose goal is making programming for parallel computing easy and pervasive. HSA members are building a heterogeneous computing ecosystem, rooted in industry standards, which combines scalar processing on the CPU with parallel processing on the GPU, while enabling high bandwidth access to memory and high application performance with low power consumption. HSA defines interfaces for parallel computation using CPU, GPU and other programmable and fixed function devices, while supporting a diverse set of high-level programming languages, and creating the foundation for next-generation, general-purpose computing.

Source: HSA Foundation

The post HSA and ROCm Architectures to be Highlighted at Next Week’s CppCon appeared first on HPCwire.

Exascale Computing Project Names Doug Kothe as Director

Wed, 09/20/2017 - 08:43

The Department of Energy’s Exascale Computing Project (ECP) has named Doug Kothe as its new director effective October 1. He replaces Paul Messina, who is stepping down after two years to return to Argonne National Laboratory.

Kothe is a 32-year veteran of DOE’s National Laboratory System. He most recently served in Oak Ridge National Laboratory’s Computing and Computational Sciences Directorate and as the applications development lead for ECP. For the preceding five years, he led the Consortium for Advanced Simulation of Light Water Reactors, DOE’s first Energy Innovation Hub, which uses supercomputers to improve nuclear reactor performance.

“Doug’s credentials in this area and familiarity with every aspect of the ECP make him the ideal person to build on the project’s strong momentum,” said Bill Goldstein, director of Lawrence Livermore National Laboratory and chairman of the ECP Board of Directors, which hired Kothe.

Kothe will be based at Oak Ridge National Laboratory, host of the ECP project office, where he will report to ORNL Director Thomas Zacharia. “Doug knows how to build strong collaborations across diverse disciplines and institutions—often among people who are typically competitors,” Zacharia said. “ECP’s success will require labs, companies, and universities to work together to accomplish something they can’t do alone: Building applications, software, and driving the hardware R&D that will enable the first U.S. exascale systems.”

Global competition is challenging U.S. leadership in high-performance computing, which has become a critical tool for accelerating solutions to problems in both science and industry. Other countries are making significant investments in both technology and research, seeking an advantage in the ongoing competition in HPC systems and software.

Exascale is the next level of performance for HPC. Today’s petascale systems are measured in quadrillions (1015) of calculations per second. Exascale systems will run at quintillions (1018) of calculations per second, more realistically simulating the processes involved in applications such as precision medicine, manufacturing, fuels and energy systems, and the nation’s stockpile stewardship program, as well as the unseen physics at work within materials and the fundamental forces of the universe. Exascale also holds tremendous potential for emerging disciplines such as large-scale data analytics, machine learning, and artificial intelligence.

The ECP was launched in 2016 as a collaboration between the DOE Office of Science and DOE’s National Nuclear Security Administration to provide exascale computing capability that is critical to DOE missions in national security, scientific discovery and economic competitiveness. The collaboration includes experts from six core national laboratories – Argonne, Lawrence Berkeley, Lawrence Livermore, Los Alamos, Oak Ridge and Sandia – along with representatives from industry and academia.

Kothe credited his predecessor’s leadership. “Paul’s management and mentoring of team members during start-up put us on a successful trajectory,” Kothe said. “My confidence going forward reflects the fact that the ECP scientists and engineers who are executing this plan are leaders in the HPC community and among the most talented in the world.”

At Argonne, Messina will return to focus on program development in the Computing, Environment, and Life Sciences Directorate and strategic computational science directions for the lab.

“I’m proud to have helped establish ECP,” Messina said. “It’s a fantastic team doing important work. I’m confident that the project will thrive in Doug’s capable hands.”

Said Goldstein, “Paul Messina’s leadership has been invaluable from day one – from establishing and building the ECP organization to strategic planning of the project’s deliverables to the execution of this critical national project over the past two years.”

Source: Department of Energy’s Exascale Computing Project

The post Exascale Computing Project Names Doug Kothe as Director appeared first on HPCwire.

Exascale Computing Project Names Doug Kothe as Director

Wed, 09/20/2017 - 07:46

OAK RIDGE, Tenn., Sept. 20 — The Department of Energy’s Exascale Computing Project (ECP) has named Doug Kothe as its new director effective October 1. He replaces Paul Messina, who is stepping down after two years to return to Argonne National Laboratory.

Doug Kothe

Kothe is a 32-year veteran of DOE’s National Laboratory System. He most recently served in Oak Ridge National Laboratory’s Computing and Computational Sciences Directorate and as the applications development lead for ECP. For the preceding five years, he led the Consortium for Advanced Simulation of Light Water Reactors, DOE’s first Energy Innovation Hub, which uses supercomputers to improve nuclear reactor performance.

“Doug’s credentials in this area and familiarity with every aspect of the ECP make him the ideal person to build on the project’s strong momentum,” said Bill Goldstein, director of Lawrence Livermore National Laboratory and chairman of the ECP Board of Directors, which hired Kothe.

Kothe will be based at Oak Ridge National Laboratory, host of the ECP project office, where he will report to ORNL Director Thomas Zacharia. “Doug knows how to build strong collaborations across diverse disciplines and institutions—often among people who are typically competitors,” Zacharia said. “ECP’s success will require labs, companies, and universities to work together to accomplish something they can’t do alone: Building applications, software, and driving the hardware R&D that will enable the first U.S. exascale systems.”

Global competition is challenging U.S. leadership in high-performance computing, which has become a critical tool for accelerating solutions to problems in both science and industry. Other countries are making significant investments in both technology and research, seeking an advantage in the ongoing competition in HPC systems and software.

Exascale is the next level of performance for HPC. Today’s petascale systems are measured in quadrillions (1015) of calculations per second. Exascale systems will run at quintillions (1018) of calculations per second, more realistically simulating the processes involved in applications such as precision medicine, manufacturing, fuels and energy systems, and the nation’s stockpile stewardship program, as well as the unseen physics at work within materials and the fundamental forces of the universe. Exascale also holds tremendous potential for emerging disciplines such as large-scale data analytics, machine learning, and artificial intelligence.

The ECP was launched in 2016 as a collaboration between the DOE Office of Science and DOE’s National Nuclear Security Administration to provide exascale computing capability that is critical to DOE missions in national security, scientific discovery and economic competitiveness. The collaboration includes experts from six core national laboratories – Argonne, Lawrence Berkeley, Lawrence Livermore, Los Alamos, Oak Ridge and Sandia – along with representatives from industry and academia.

Kothe credited his predecessor’s leadership. “Paul’s management and mentoring of team members during start-up put us on a successful trajectory,” Kothe said. “My confidence going forward reflects the fact that the ECP scientists and engineers who are executing this plan are leaders in the HPC community and among the most talented in the world.”

At Argonne, Messina will return to focus on program development in the Computing, Environment, and Life Sciences Directorate and strategic computational science directions for the lab.

“I’m proud to have helped establish ECP,” Messina said. “It’s a fantastic team doing important work. I’m confident that the project will thrive in Doug’s capable hands.”

Said Goldstein, “Paul Messina’s leadership has been invaluable from day one – from establishing and building the ECP organization to strategic planning of the project’s deliverables to the execution of this critical national project over the past two years.”

Source: Department of Energy’s Exascale Computing Project

 

The post Exascale Computing Project Names Doug Kothe as Director appeared first on HPCwire.

Secretary of Energy Rick Perry Announces High Performance Computing for Materials Program

Tue, 09/19/2017 - 19:00

WASHINGTON, D.C., Sept. 19 – Today, U.S. Secretary of Energy Rick Perry (DOE) announced a new high-performance computing (HPC) initiative that will help U.S. industry accelerate the development of new or improved materials for use in severe environments.

“The High Performance Computing for Materials Program will provide opportunities for our industry partners to access the high-performance computing capabilities and expertise of DOE’s national labs as they work to create and improve technologies that combat extreme conditions,” said Secretary Perry. “This initiative combines two, crucial elements of the Administration’s mission at DOE – advances in high-performance computing and the improved transition of energy technologies to market.”

The HPC4Mtls initiative will initially focus on challenges facing industry as they work to develop new or improved materials that can sustain extreme conditions—including extreme pressure, radiation, and temperature, corrosion, chemical environment, vibration, fatigue, or stress states. It will focus on developing improved lightweight material technologies, as well. The program aims to enable a step change in the cost, development time, and performance of materials in severe environments and save millions of dollars in fuel and maintenance across sectors. These material advancements will also increase U.S. competitiveness in the global marketplace.

Through HPC4Mtls, industry will be able to solve common materials issues, discover new or improved materials and structures, and enhance their products and processes using the labs’ world-class computational resources and capabilities. These capabilities include:

  • Access to HPC systems, including five of the world’s ten fastest computers
  • Higher-fidelity simulations to augment products or processes
  • Prediction of material behavior in specific severe environments
  • Modeling of missing physical phenomena to enable more realistic simulations
  • Development of more complex models to capture interactions between physical phenomena
  • Access to expertise in computational fluid dynamics, thermodynamics, kinetics, materials modeling, and additive manufacturing.

Companies will be selected to participate in the initiative through an open, two-stage, competitive process and will contribute at least 20 percent of project costs. DOE will hold a closed-press workshop on October 12, 2017, in Pittsburgh, Pa., to provide more information on the program and engage U.S.-based companies, industry, universities, and government stakeholders.

Sponsored by DOE’s Office of Fossil Energy, the High Performance Computing for Materials (HPC4Mtls) Program is part of the larger HPC4 Energy Innovation Initiative, a Department-wide effort comprised of the Office of Fossil Energy, the Office of Energy Efficiency and Renewable Energy, and the Office of Nuclear EnergyLawrence Livermore National LaboratoryLos Alamos National LaboratoryOak Ridge National Laboratory, and the National Energy Technology Laboratory serve as the principal leads on this initiative, which could ultimately lower emissions, reduce fuel and maintenance costs across the economy, and save millions of dollars.

For more information about the HPC4Mtls program and to register for the Oct. 12th workshop, please click HERE.

Source: US Department of Energy

The post Secretary of Energy Rick Perry Announces High Performance Computing for Materials Program appeared first on HPCwire.

Takeaways from the Milwaukee HPC User Forum

Tue, 09/19/2017 - 18:46

Milwaukee’s elegant Pfister Hotel hosted approximately 100 attendees for the 66th HPC User Forum (September 5-7, 2017). In the original home city of Pabst Blue Ribbon and Harley Davidson motorcycles the agenda addressed artificial intelligence, self-driving vehicles and drug repositioning. The printed agenda neglected to suggest that we would actually be served PBR and be accompanied by two HD cruisers complements of the House of Harley local dealership. The Hyperion folks surprised and delighted us further with live Germany music, rouladen, bratwurst and sauerkraut.

Below are a few of my observations from the forum.

Exascale Efforts in the USA and EU

While the European Union’s single digital market strategy is moving forward with a legal and procurement framework, the USA is thinking through metrics. Specifically, it appears the exascale community has abandoned metrics for theoretical peak performance and the percentage utilization of a CPU as key metrics. This may portend that less attention is being given to solver-heavy physics to this new generation of supercomputers. We shall see, but perhaps this is influenced by the fraction of non-recurring engineering costs involved in developing exascale systems in a non-incremental way. Seemingly arbitrary power limitations and an observed pullback on metrics in the US model may correlate with some community observations of under-investment. But perhaps an approach that doesn’t require as much double precision will broaden the market.

Key takeaway: Exascale will emerge in unexpected ways following a retrenchment in HPC metrics used for decades

Artificial Intelligence/Machine Learning/Deep Learning

There were several presentations on AI, machine learning and deep learning ranging from Michael Garris who is co-chair of the NIST ML/AI subcommittee to Maarten Sierhuis (Nissan Research Center in Silicon Valley), Tim Barr (Cray), Arno Kolster (Providentia Worldwide) and Graham Anthony (BioVista). While each of us knows intuitively that we have cognitive assistance in our pockets I was especially interested in the comments that accuracy and speed is often a tradeoff (logical), reduction in error rates occur when 10x more data is used (nice quantification) and pattern detection is very specific to the use case (less intuitive).

Maarten Sierhuis predicted that multiple lane highway scans for automobiles will be available in 2018 and for urban intersections in 2020. Full autonomy is extremely difficult, especially when attempting to identify non-car objects and mimic human decision making in complex situations. High-definition maps aren’t the only missing piece – AI must be present in the cloud.

Arno Kolster was especially targeted in his message that interoperability and workflow management lag pattern detection and algorithm development, concluding that general solutions are a long way off. Algorithm and data formats are very closely linked now – a lockstep that is predictable but inflexible. Ideally, algorithm performance would detect and adjust to system capabilities, along with fluid workflow, integrated message flow, visualization tuned to the customer and well exposed KPIs.

A breath of fresh air came from Graham Anthony who spoke about the pursuit of sustainable healthcare through personalized medicine. The BioVista website calls it ‘drug repositioning’ when HPC drives the ability to more effectively and quickly combine patient and general biomedical data to transform medicine. The key challenge is to get the cost of these services down to fit into standard cost reimbursement codes and the time-frame for doctor use to fit into a 15-minute visit.

Key takeaway: High qualitative impact in a variety of sectors may dwarf the use of today’s research HPC

Innovation Award Winners Paul Morin (The Polar Geospatial Center University of Minnesota) and Leigh Orf (University of Wisconsin at Madison)

Dr. Morin caught my attention when he claimed he could use all possible cycles in the world to analyze geospatial mapping of the poles. Perhaps he said he could use all cycles ever provided but I got rather lost in just the current realm. His plan is to process 80 trillion pictures of the entire arctic at a resolution of two meters. Then repeat – effectively providing time-dependent photography that can track changes in elevation. He uses Blue Waters as a capacity machine today but its scheduler had to be rewritten to handle thousands of job launches. My first thought was that other use cases could benefit from a high-capacity scheduler, such as bioinformatics. Then my second thought was a bit cynical, thinking that most capacity computing proposals wither and die among policy makers who believe our nation’s largest machines should be reserved for capability computing. He is willing to try other technologies. Perhaps the cloud’s existing exascale capacity could help – its current business model notwithstanding.

Dr. Orf’s tornadoes were among the best I’ve seen. He uses 15-meter resolutions knowing that doubling the resolution needs 10x more compute power and 8x more memory. His biggest bottleneck is I/O because of the frequency of time-step saving. His biggest achievement may be that he effectively created a new file system by allocating one core per node to build an HDF5 file. His key desire is to issue probabilistic forecast warnings by looking at radar as storms are forming and differentiating between predictions of EF1 and EF5.

Key takeaway: These researchers are heroes interested in impact that transcends both basic and applied research. So why is ready access to huge, highly-tuned capacity computing so impossible?

NCSA/Hyperion Industry Study

I have some unique perspectives on the report released August 22, 2017, by NCSA since I was the initial PI for the NSF award. This work complements a 2012 NCSA survey that I completed on the impact on scientific discovery using simulation-based engineering and science and a 2015 book on industrial applications of HPC that captures 40 contributions from eleven countries from HPC centers that engage closely with industry. I’ll share my observations on this study for a separate article.

Merle Giles (NCSA Private Sector Program – U.S.)

As for my takeaways from beyond the printed agenda I would simply observe that the dinner speaker from the Pabst Museum was informative and inspirational. Captain Pabst married into a brewing family and became an unlikely company president given his first love as a steamer captain on Lake Michigan. Pabst Brewing Company ultimately grew to become the world’s largest brewery, selling 15.6 million barrels of beer in 1978. I highly recommend a tour of the 22,000 square-foot 1890s-era Pabst mansion on Milwaukee’s original Grand Avenue. It offers deep learning of a different kind.

About the Author

Merle Giles is currently CEO of Moonshot Research LLC. He directed NCSA’s Private Sector Program for ten years.

The post Takeaways from the Milwaukee HPC User Forum appeared first on HPCwire.

NSF Awards $10M to Extend Chameleon Cloud Testbed Project

Tue, 09/19/2017 - 12:49

The National Science Foundation has awarded a second phase, $10 million grant to the Chameleon cloud computing testbed project led by University of Chicago with partners at the Texas Advanced Computing Center (TACC), Renaissance Computing Institute (RENCI), and Northwestern University. The three-year grant will add hardware and expand access.

Currently Chameleon is approximately 600-node cloud infrastructure with bare metal reconfiguration privileges. This level of access allows researchers to go beyond limited development on existing commercial or scientific clouds, offering a customizable platform to create and test new cloud computing architectures.

“In phase one we built a testbed, but in phase two we’re going to transform this testbed into a scientific instrument,” said Kate Keahey, Argonne computer scientist, Computation Institute fellow, and Chameleon project PI. “We’re going to extend the capabilities that allow users to keep a record of their experiments in Chameleon and provide new services that allow them to build more repeatable experiments.”

In its first phase, Chameleon supported work in computer science areas such as cybersecurity, OS design and power management; researchers could “realistically simulate cyberattacks upon cloud computing systems to improve their defenses, train students to search high-resolution telescope images for undiscovered exoplanets, and develop machine learning algorithms that automatically determine the most energy-efficient task assignment schemes for large data centers.”

Phase two includes adding racks at UChicago and TACC, infusion of highly-contested resources such as GPUs, and Corsa network switches. The new Corsa switches enable experimentation with software-defined networking (SDN) within a Chameleon site as well as extending individual SDN experiments across the wide-area to include resources from either Chameleon site or even from other compatible testbeds, such as NSF’s GENI.

On the software side, the Chameleon team will package CHI (CHameleon Infrastructure), the software operating Chameleon, based primarily on the open-source OpenStack project to which the University of Chicago team made substantial contributions. Packaging the Chameleon operational model will allow others to create their own experimental clouds easily.

Link to TACC article: https://www.tacc.utexas.edu/-/cloud-computing-testbed-chameleon-renewed-for-second-phase

The post NSF Awards $10M to Extend Chameleon Cloud Testbed Project appeared first on HPCwire.

ORNL Innovation Crossroads Program Opens Second Round of Energy Entrepreneurial Fellowships

Tue, 09/19/2017 - 10:47

OAK RIDGE, Tenn., Sept. 19, 2017 — Entrepreneurs are invited to apply for the second round of Oak Ridge National Laboratory’s Innovation Crossroads program.

Up to five innovators will receive a two-year post-doctoral entrepreneurial fellowship that includes vouchers worth up to $350,000 for collaborative research and development at ORNL, startup business guidance and mentoring, and health and travel benefits. Entrepreneurs selected during the merit-based process are expected to begin the program by May 2018.

Innovation Crossroads is one of three U.S. Department of Energy Lab-Embedded Entrepreneurship Programs designed to embed top technical post-doctoral talent within national labs as research fellows with goal of subsequently launching businesses. The program focuses on early-stage research and development along with entrepreneurial guidance to enable innovators to inject new ideas into the national labs and transform their novel ideas into U.S.-based companies.

Innovation Crossroads fellows have access to world-class research facilities and scientific expertise at ORNL, including the Manufacturing Demonstration Facility, the National Transportation Research Center, the Oak Ridge Leadership Computing Facility, the Center for Nanophase Materials Sciences, and the Spallation Neutron Source.

Through regional partnerships with entrepreneurial and business accelerator organizations, fellows also receive assistance with developing business strategies, conducting market research, introductions to potential commercial partners, and finding additional sources of funding.

The first cohort of Innovation Crossroads fellows began their terms this summer. These first-time entrepreneurs include Anna Douglas (SkyNano), who is developing a process that uses carbon dioxide as a feedstock to produce low-cost carbon nanotubes; Matthew Ellis and Samuel Shaner (Yellowstone Energy), who are jointly developing an advanced nuclear reactor design; and Mitchell Ishmael (Active Energy Systems), who is developing a system for low level heat recovery and energy storage.

Interested entrepreneurs can learn about Innovation Crossroads and begin the application process at innovationcrossroads.ornl.gov. The application period closes Oct. 30.

Innovation Crossroads is funded by DOE’s Advanced Manufacturing Office (AMO), which supports early-stage applied research and development of new materials, information, and process technologies that improve American manufacturing’s energy efficiency, as well as platform technologies for manufacturing clean energy products.

EERE supports early-stage research and development of energy efficiency and renewable energy technologies that make energy more affordable and strengthen the reliability, resilience, and security of the U.S. electric grid.

UT-Battelle manages ORNL for the DOE’s Office of Science. The DOE Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time. For more information, visit the Office of Science website.

Source: ORNL

The post ORNL Innovation Crossroads Program Opens Second Round of Energy Entrepreneurial Fellowships appeared first on HPCwire.

Intel Custom Foundry Certifies ANSYS Simulation Tools for 22 nm FinFET Low-power Process Technology

Tue, 09/19/2017 - 10:40

PITTSBURGH, Sept. 19, 2017 — Intel Custom Foundry customers are delivering powerful, innovative products thanks to the certification of ANSYS (NASDAQ: ANSS) solutions for electromigration, power and electrostatic discharge reference flows for Intel 22nm FinFET low-power (22FFL) process technology. The supported tools from ANSYS and Intel Custom Foundry’s collaboration enable mutual customers to minimize design costs and risks while quickly bringing cutting-edge and reliable products to market.

IoT and entry mobile products demand higher performance with less power consumption and excellent reliability. To achieve this today, multiple subsystems of an electronic product are combined into one or more integrated circuits known as a system on a chip (SoC). ANSYS simulation tools deliver needed accuracy while reducing turnaround time to meet the increased computational requirements caused by the growing design complexity of modern products. Advanced technology support in ANSYS RedHawkANSYS Totem and ANSYS PathFinder, including electromigration rule compliance, deliver greater reliability and manufacturability, as well as minimize risk and lower cost.

Intel’s new 22FFL process technology offers a unique blend of high-performance and ultralow-power transistors combined with simplified interconnects and simpler design rules to deliver a versatile FinFET design platform for low-power and mobile products. It offers up to 100x lower leakage compared with the previous Intel 22GP (general purpose) technology. The Intel 22FFL also delivers drive currents on par with Intel’s 14 nm transistors while delivering true 22nm class scaling at 17.8 MTr/mm^2, enabling better performance and power than any industry planar technology can achieve.

The certification from Intel Custom Foundry for its advanced 22FFL process technology validates the capability to simulate designs while maintaining sign-off accuracy. It enables designers to meet demanding power and reliability requirements for their intellectual properties, analog and custom integrated circuit designs. Mutual customers of Intel Custom Foundry and ANSYS can design cutting-edge applications such as mobile and low power IoT products based on this 22FFL certification.

“22FFL is a unique new technology that provides a compelling combination of performance, power, density and ease of design for low-power IoT and mobile products,” said Venkat Immaneni, senior director, Foundry Design Kit Enablement for Intel Custom Foundry. “The certification of ANSYS tools combined with the comprehensive Intel Custom Foundry 22FFL platform gives our mutual customers a competitive advantage when implementing robust, high-performance intellectual properties and SoCs on our new 22FFL offerings.”

“Power, electromigration and electrostatic discharge reference flows are absolute requirements to create smart, robust IoT and entry mobile products. Our collaboration with Intel Custom Foundry on the 22FFL design platform and its certification of ANSYS solutions emphasize the high-quality results and benefits of ANSYS simulation tools,” said John Lee, general manager at ANSYS. “This collaborative effort further empowers Intel Custom Foundry customers to confidently build the next-generation of robust and reliable computing products for low-power IoT and mobile products.”

About ANSYS, Inc.

If you’ve ever seen a rocket launch, flown on an airplane, driven a car, used a computer, touched a mobile device, crossed a bridge, or put on wearable technology, chances are you’ve used a product where ANSYS software played a critical role in its creation. ANSYS is the global leader in Pervasive Engineering Simulation. We help the world’s most innovative companies deliver radically better products to their customers. By offering the best and broadest portfolio of engineering simulation software, we help them solve the most complex design challenges and create products limited only by imagination. Founded in 1970, ANSYS employs thousands of professionals, many of whom are expert M.S. and Ph.D.-level engineers in finite element analysis, computational fluid dynamics, electronics, semiconductors, embedded software and design optimization. Headquartered south of Pittsburgh, Pennsylvania, U.S.A., ANSYS has more than 75 strategic sales locations throughout the world with a network of channel partners in 40+ countries. Visit www.ansys.com for more information.

Source: ANSYS, Inc.

The post Intel Custom Foundry Certifies ANSYS Simulation Tools for 22 nm FinFET Low-power Process Technology appeared first on HPCwire.

NERSC Simulations Shed Light on Fusion Reaction Turbulence

Tue, 09/19/2017 - 10:03

Understanding fusion reactions in detail – particularly plasma turbulence – is critical to the effort to bring fusion power to reality. Recent work including roughly 70 million hours of compute time at the National Energy Research Scientific Computing Center (NERSC) is helping clarify turbulence behavior characteristics.

A team of physicists from the University of California at San Diego (UCSD), MIT’s Plasma Science and Fusion Center and Princeton Plasma Physics Laboratory (PPPL) ran a series of multiscale gyrokinetic simulations at Lawrence Berkeley National Laboratory’s National Energy Research Scientific Computing Center (NERSC) to determine whether electron energy transport in a tokamak plasma discharge is multiscale in nature.

Visualization of temperature fluctuations from a high-resolution simulation of a plasma discharge in the DIII-D tokamak. The DIII-D plasma was designed to match many of the plasma parameters targeted in ITER operation. Image: Chris Holland

The simulation added strong evidence that electron energy transport is indeed a multi-scale phenomenon. Being able to accurately predict electron energy transport is critical for predicting performance in future reactors such as ITER, a global collaboration to build the largest tokomak fusion reactor, currently under construction in Cadarache, France.

“In a fusion reactor, most of the heat generated in the plasma will be transported by the electrons,” said Chris Holland, a research scientist in the Center for Energy Research at UCSD and lead author on a recent study in Nuclear Fusion describing this work. This study builds on previous research by Holland and colleagues at MIT and General Atomics in which they used multiscale simulations to more precisely study the turbulence instabilities that cause plasma heat loss.

These latest simulations, which were performed with the GYRO gyrokinetic plasma turbulence code and used nearly 70 million hours of computing time on NERSC’s Edison system, corresponded to conditions measured in a plasma run at the DIII-D tokamak reactor using the ITER baseline scenario. Edison is a Cray XC30, with a peak performance of 2.57 petaflops/sec, 133,824 compute cores, 357 terabytes of memory, and 7.56 petabytes of disk.

Link to paper on the work: Gyrokinetic predictions of multiscale transport in a DIII-D ITER baseline discharge

Link to NERSC article: Multiscale Simulations Help Predict Unruly Plasma Behavior

Feature image: fusion reactor

The post NERSC Simulations Shed Light on Fusion Reaction Turbulence appeared first on HPCwire.

JAIST Puts Cray XC40 Supercomputer into Production

Tue, 09/19/2017 - 09:16

SEATTLE, Sept. 18, 2017 — Supercomputer leader Cray Inc. (Nasdaq: CRAY) today announced the Japan Advanced Institute for Science and Technology (JAIST) has put a Cray XC40 supercomputer into production. JAIST, a postgraduate university located in Nomi, Ishikawa, Japan and one of the country’s premier academic research centers, is using its new Cray XC40 system as it primary supercomputing resource supporting computational research across the University.

The three-cabinet Cray XC40 supercomputer is the latest in a long history of Cray systems used to advance scientific research at JAIST. Previous Cray supercomputers at JAIST have included the Cray T3E, Cray XT3, and Cray XC30 systems. In addition to powering data-intensive research in a wide array of scientific disciplines at JAIST, the new Cray XC40 supercomputer will also help speed advancements in the development of new algorithms for highly-parallel computers and will perform large-scale simulations in nanotechnology and biomechanics.

“JAIST is committed to conducting world-class research activities that leverage the abilities of our highly-skilled faculty members and students, our advanced research facilities, and cutting-edge technologies, and Cray supercomputers continue to play a vital role in our efforts,” said Yasushi Inoguchi, Professor of the Research Center for Advanced Computing Infrastructure, JAIST. “Our new Cray XC40 supercomputer will support our mission of becoming a premier center of excellence in education and research.”

“Cray has enjoyed a long, collaborative partnership with JAIST, and we continue to take great pride in providing the University’s user community with advanced supercomputing technologies for achieving breakthrough results,” said Mamoru Nakano, president of Cray Japan. “The addition of this new Cray system at the JAIST is another example of our continued drive to expand our presence in Japan.”

The Cray XC series of supercomputers are designed to handle the most challenging workloads requiring sustained multi-petaflop performance. The Cray XC40 supercomputers incorporate the Aries high performance network interconnect for low latency and scalable global bandwidth, as well as the latest Intel Xeon processors, Intel Xeon Phi processors, and NVIDIA Tesla GPU accelerators. The Cray XC supercomputers deliver on Cray’s commitment to performance supercomputing with an architecture and software environment that provides extreme scalability and sustained performance.

For more information on the Cray XC supercomputers, please visit the Cray website at www.cray.com.

About JAIST

Japan Advanced Institute of Science and Technology (JAIST) was founded in October 1990 as the first independent national graduate school in Japan to carry out graduate education based on research at the highest level in advanced science and technology. Organized into three graduate schools (knowledge science, information science and materials science), JAIST provides up-to-date graduate training for about 900 students while maintaining competitive research programs through various dedicated centers, such as technical computing, nanotechnology and information security. For more information, go to www.jaist.ac.jp.

About Cray Inc.

Supercomputing leader Cray Inc. (Nasdaq: CRAY) provides innovative systems and solutions enabling scientists and engineers in industry, academia and government to meet existing and future simulation and analytics challenges. Leveraging more than 40 years of experience in developing and servicing the world’s most advanced supercomputers, Cray offers a comprehensive portfolio of supercomputers and big data storage and analytics solutions delivering unrivaled performance, efficiency and scalability. Cray’s Adaptive Supercomputing vision is focused on delivering innovative next-generation products that integrate diverse processing technologies into a unified architecture, allowing customers to meet the market’s continued demand for realized performance. Go to www.cray.com for more information.

Source: Cray

The post JAIST Puts Cray XC40 Supercomputer into Production appeared first on HPCwire.

High Schoolers Sharpen Civil Engineering and Computation Skills at Code@TACC Summer Camp

Tue, 09/19/2017 - 09:04

Sept. 19, 2017 — What if high school kids could make a building stand up to an earthquake? A summer camp at the Texas Advanced Computing Center (TACC) smoothed the way for students to learn about the science behind building design for earthquakes.

Thirty high school students from Texas and Louisiana sharpened their skills in civil engineering and computation at a 2017 summer camp called Code@TACC DesignSafe. The camp was supported by DesignSafe, a national cyberinfrastructure program funded by the National Science Foundation. DesignSafe is a web-based research platform of the Natural Hazards Engineering Research Infrastructure Network (NHERI). It helps engineers perform research that leads to building safer structures that are resilient to natural hazards such as earthquakes, windstorms, and hurricanes.

Code@TACC DesignSafe gave the students a taste of what it’s like to engineer for earthquakes. The students made buildings out of the construction toy K’NEX. They tested these structures at the UT Austin Ferguson Structural Engineering Lab, where they simulated the effects of an earthquake. They used a shake table, a computer-controlled motorized table that can re-create the wave patterns of historically significant earthquakes. The students collected movement data via accelerometers attached to their model buildings. They analyzed the data with the Python programming language and looked for resonant frequencies of the structures.

Student teams presented their findings after four days of building, testing, data collection, and making videos of their structures shaking in the lab. Ultimately, they were able to evaluate how well their buildings responded to different types of earthquakes based on historic earthquake data.

The student teams studied a handful of different earthquakes from history. The 2003 laterally-moving strike-slip quake in Bam, Iran, gave them a vivid example of disaster: an estimated 40,000 people killed and 20,000 more injured.

“We believe a reason for this high casualty rate is because of the structures of the building,” said camp student Briana Cuero at her team’s final presentation as she pointed at slides showing the aftermath. “You can see in the image of the buildings in Bam, they’re mostly made of cement from clay that is ample in this region. We decided to create a building which could sustain its structure and wouldn’t collapse drastically.”

She explained that the diagonal struts would help stabilize the walls of her team’s model. “It also has a little star for a decoration,” Briana added.

Student Damarius Kennedy reminded everyone about the limitations facing their structures. They had to be at least 18 inches tall and couldn’t be wider than the boards selected to sit on the shake tables. They were confined to a budget of 10 dollars, which Kennedy’s team spent on balsa wood, hot glue, and gorilla tape to stabilize the roof and struts. What’s more, they could only make small changes to their design the first rounds from their nemesis, the shake table.

“One issue we had is with the weights on the structures. They bowed,” Damarius pointed out. “The green pieces fell out, and our roof collapsed. Our solution was to put ties on the place where it collapsed, connecting to the lower green and the blue connectors of the K’NEX,” she said.

Student Etienne Cuero showed the audience of parents, students, and TACC staff their data analysis. “We used graphs to determine the amplitude and the frequency of each earthquake test,” Etienne explained.

Each team also had to discuss ways to improve their results. “If we could do this again, we could add supports on all four walls,” said team member Max Irby. “On the second story, we could add that support and make everything more equal and more stable.”

Max added that they would also remove the weight from the top. “That star didn’t serve any purpose, and we had already met the height requirement,” he said.

“In conclusion,” said Etienne,”our building moved just as much as the earthquake and a bit more, which is bad because if you were in that situation you would have unpredictable movement. And things landing on you hurts!”

“This is a fantastic program,” said Code@TACC DesignSafe instructor Chunxiao Ge (Emma Gee), a physics and biology teacher at the Colorado River Collegiate Academy of Bastrop ISD. “Kids spend four days and have a deep understanding of the basic physics and math. Here in this camp, students interpret data and graphs in a way that is true to the real world,” Ge said.

Code@TACC DesignSafe instructor Patty Hill, an algebra teacher at Kealing Middle School of Austin ISD, agreed that the camp went well. “What this project does, and what this camp did, was it brought things together in a way where everything made sense. You see how science and math and the real-world experiences of the civil engineer, and the aftermath of the earthquakes — it all blended together,” Hill said.

Hill and Ge participated in the DesignSafe Research Experience for Teachers, where they learned new ways to apply coding and analysis and how to teach engineering through Jupyter notebooks and the Python programming language. After the camp, the teachers were given K’NEX building supplies; materials to build their own classroom shake table; and access to the DesignSafe web portal.

“We want the students to be able to bring back to their schools the same type of engineering experiences so that students who can’t attend the summer camp can also have those same type of experiences in the classroom,” Joon-Yee Chuah of TACC said.

“We want students and parents to know that coding is going to be a fundamental part of any engineering or scientific field in the future,” Chua continued. “Maybe they aren’t interested in things like programming robots or programming apps and games. But coding is still going to be important. We want to show students that they can both have a hands on experience doing things like building structures, and then still use coding as part of those engineering projects. So it’s truly multidisciplinary,” Chuah said.

Original article: https://www.tacc.utexas.edu/-/buildings-vs-earthquakes-high-school-students-learn-the-science 

Source: Jorge Salazar, Texas Advanced Computing Center

The post High Schoolers Sharpen Civil Engineering and Computation Skills at Code@TACC Summer Camp appeared first on HPCwire.

Bright Computing Announces Support for Ubuntu

Tue, 09/19/2017 - 08:24

SAN JOSE, Calif., Sept. 19, 2017 –Bright Computing, a global leader in cluster and cloud infrastructure automation software, today announced the general availability of Bright Cluster Manager 8.0 with Ubuntu.

With this integration, organizations can run Bright Cluster Manager Version 8.0 on top of Ubuntu, to easily build, provision, monitor and manage Ubuntu high performance clusters from a single point of control, in both on-premises and cloud-based environments.

Ubuntu is an open-source platform for client, server and cloud computing and has become a natural choice for users of all kinds, from Fortune 500 companies to hardware makers, content providers, software developers and individual technologists.  Canonical is the commercial sponsor of the Ubuntu project and the leading provider of support services for Ubuntu deployments in the enterprise.

Bright Computing is well known for its powerful infrastructure management technology. The company is also developing a reputation in the deep learning space, as Bright Cluster Manager makes it faster and easier for organizations to stand up deep learning platforms to gain actionable insights from the rich, complex data associated with machine learning and artificial intelligence. As a result, a growing number of Bright customers are extending their use of Bright to manage not just their clustered infrastructure, but their deep learning environment as well. All of Bright’s deep learning packages are now available for Ubuntu, one of the world’s most popular deep learning Linux platforms. The integration between Bright and Ubuntu therefore offers great benefit to organizations with a deep learning requirement.

Martijn de Vries, CTO at Bright Computing, commented; “Many of our customers use Ubuntu and have been eager for Bright to integrate Bright Cluster Manager and Ubuntu. We are pleased that we can now satisfy this requirement with an enterprise-grade solution.”

Udi Nachmany, VP of Cloud Alliances at Canonical, said: “Ubuntu runs most public cloud and OpenStack workloads, and is rapidly gaining ground in on-prem, bare metal environments. This is because our licensing model supports scale-out, while our distribution remains truly enterprise-grade. With this integration, Bright Computing users will now have access to a reliable, production-ready Ubuntu straight from the source, and with access to enterprise support with all of the security, uptime and tooling that entails.”

About Bright Computing

Bright Computing is the leading provider of hardware-agnostic cluster and cloud management software in the world. Bright Cluster Manager, Bright Cluster Manager for Big Data, and Bright OpenStack provide a unified approach to installing, provisioning, configuring, managing, and monitoring HPC clusters, big data clusters, and OpenStack clouds. Bright’s products are currently deployed in more than 650 data centers around the world. Bright Computing’s customer base includes global academic, governmental, financial, healthcare, manufacturing, oil/gas/energy, and pharmaceutical organizations such as Boeing, Intel, NASA, Stanford University, and St. Jude Children’s Research Hospital. Bright partners with Amazon, Cray, Dell, Intel, Nvidia, SGI, and other leading vendors to deliver powerful, integrated solutions for managing advanced IT infrastructure such as high-performance computing clusters, big data clusters, and OpenStack-based private clouds.

Source: Bright Computing

The post Bright Computing Announces Support for Ubuntu appeared first on HPCwire.

Mellanox Appoints Shlomit Weiss to Senior Vice President of Silicon Engineering

Mon, 09/18/2017 - 07:32

SUNNYVALE, Calif. & YOKNEAM, Israel, Sept. 18, 2017 — Mellanox Technologies, Ltd. (NASDAQ:MLNX), a leading supplier of high-performance, end-to-end smart interconnect solutions for data center servers and storage systems, today announced the appointment of Shlomit Weiss to senior vice president of silicon engineering. Ms. Weiss will report to Eyal Waldman, Mellanox’s president and CEO.

“I am very pleased to have Shlomit join the Mellanox leadership team, bringing her vast experience, track record and extensive knowledge to lead our silicon engineering team,” said Eyal Waldman, president, CEO of Mellanox Technologies. “We look forward to Shlomit’s support in scaling Mellanox to our next stages of growth. I would like to thank Alon Webman who managed the silicon engineering team to-date, as he continues to help with the execution of the company missions.”

Shlomit Weiss has served as vice president of silicon development and held management roles such as client system-on-a-chip design manager and various server components at Intel since 1989. Ms. Weiss holds an MSc in electrical engineering (1989, Cum Laude) and BSc (1987) in computer science from the Technion – Israel Institute of Technology. Ms. Weiss holds multiple patents in the field of microprocessor design, and has received Intel Achievement Award for her contributions.

About Mellanox

Mellanox Technologies (NASDAQ: MLNX) is a leading supplier of end-to-end InfiniBand and Ethernet smart interconnect solutions and services for servers and storage. Mellanox interconnect solutions increase data center efficiency by providing the highest throughput and lowest latency, delivering data faster to applications and unlocking system performance capability. Mellanox offers a choice of fast interconnect products: adapters, switches, cables and transceivers, software and silicon that accelerate application runtime and maximize business results for a wide range of markets including high performance computing, enterprise data centers, Web 2.0, cloud, storage and financial services. More information is available at: www.mellanox.com.

Source: Mellanox

The post Mellanox Appoints Shlomit Weiss to Senior Vice President of Silicon Engineering appeared first on HPCwire.

AI: Deeper Learning with Intel® Omni-Path Architecture

Mon, 09/18/2017 - 01:01

Deep learning is a powerful tool that identifies patterns, extracts meaning from large, diverse datasets, and solves complex problems.  However, integrating neural networks into existing compute environments is a challenge that often requires specialized and costly infrastructure.

New software and hardware options will simplify the complexity.

  • Figure 1. A 32-node cluster based on Intel® Xeon Phi™ processors and Intel® Omni-Path Architecture demonstrated near-linear scaling running a neural network training workload based on Google TensorFlow.

    Intel® Omni-Path Architecture (Intel® OPA) is well suited to the demands of deep learning, enabling near-linear scalability across large numbers of nodes to provide fast time to results for large problems (see Figure 1).

  • Intel® Xeon® Scalable processors provide up to 2.2X higher neural network training performance than previous-generation Intel Xeon processors.[i]
  • Intel® Xeon Phi™ processors provide extreme parallelism, and deliver up to a teraflop or more of performance for neural network training, without the inherent latencies of GPUs or other PCIe-connected devices.
  • Intel optimized tools, libraries, and frameworks for deep learning provide better performance on Intel architecture than non-optimized software.

A key focus of deep learning implementations is to reduce the time to train the model and to ensure a high level of accuracy. HPC clusters provide a scalable foundation for addressing this need.[ii] However, due to workload characteristics and the compute capabilities of Intel Xeon processors, a high speed, low latency network fabric interconnect is needed to reduce the chance of a performance bottleneck. The fabric must allow all nodes to communicate quickly and effectively, so the servers don’t waste valuable compute cycles waiting to send and receive information.

As part of Intel® Scalable System Framework (Intel® SSF), Intel OPA is designed to tackle the compute- and data-intensive workloads of deep learning and other HPC applications. This high-speed fabric is developed in tandem with Intel compute and storage technologies. The resulting integration helps to resolve many of the performance and cost challenges associated with traditional HPC fabrics.

A Fabric for the Future of AI—and Other HPC Workloads

Deep learning frameworks differ, but the general workflow is the same as it is for many other HPC applications: work the calculation, iterate, then blast out the results to adjacent workloads. During the data sharing stage, a high volume of very small, latency-sensitive messages is broadcast across the fabric.

Breaking Down Barriers in AI

As the interconnect for the Pittsburgh Supercomputing Center’s supercomputer, known as Bridges, Intel® Omni-Path Architecture (Intel® OPA) is already helping to push the boundaries of AI. Bridges compute resources were used to train and run Libratus, an AI application that beat four of the world’s top poker players in a no-limit, Texas Hold ‘em tournament.

The performance and scale of Bridges enabled Libratus to refine its strategy each night based on the previous day’s play. One player said it felt like he was “playing against someone who could see his cards.”

The victory was about more than bragging rights. Libratus is applicable to other two-player zero-sum games, such as cyber-security, adversarial negotiations, and military planning, so beating humans has profound implications.

Read more about the Bridges supercomputer and Intel OPA.

Intel OPA transmits this traffic with the same 100 Gbps line speed as other high-speed fabrics, but this tells only part of the story. It also includes optimizations that address common bottlenecks.

  • Low-Latency, Even at Extreme Scale. Intel OPA provides traffic shaping and quality of service features to improve data flow and prioritize MPI traffic. These advantages help to reduce latency by up to 11 percent versus EDR InfiniBand, with up to 64 percent higher messaging rates.[i]
  • Better Price Performance. Intel OPA is based on a 48-port chip architecture (versus 36-port for InfiniBand). This reduces the number of switches, cables, and switch hops in medium to large clusters, which provides both cost and performance advantages.
  • Improved Accuracy and Resilience. Unlike InfiniBand, Intel OPA implements no-latency error checking, which improves data accuracy without slowing performance. It also stays up and running in the event of a physical link failure, so applications can run to completion, a crucial advantage for lengthy training runs.
Tight Integration Throughout the Stack

Tight integration among Intel OPA and the other components defined by Intel SSF provides additional value. For example, Intel Xeon Scalable processors and Intel Xeon Phi processors are available with integrated Intel OPA controllers to reduce the cost associated with separate fabric cards.

Intel also developed and tested Intel OPA in combination with our full HPC software stack, including Intel® HPC Orchestrator, Intel® MPI, the Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN), and the Intel® Machine Learning Scaling Library (Intel® MLSL). This integration helps to improve performance and reliability. It also reduces the complexity of designing, deploying, and managing an HPC cluster.

Figure 2. The Intel® Scalable System Framework simplifies the design of efficient, high-performing clusters that optimize the value of HPC investments. A Faster Road to Pervasive Intelligence

Learn more about Intel SSF benefits for AI and other HPC workloads at each level of the solution stack: compute, memory, storage, fabric, and software.AI is still in its infancy. Tomorrow’s neural networks will dwarf those of today. The mission of Intel OPA and the full Intel SSF solution stack is to make the computing foundation for this growth as simple, scalable and affordable as possible, not only for AI, but for all HPC workloads. This will help to ensure that front-line innovators have the tools they need to support their core mission—transforming the world through deep, pervasive intelligence.

[1] For details, see https://www.intel.com/content/www/us/en/processors/xeon/scalable/xeon-scalable-platform.html

[2] Not all deep learning frameworks are optimized to run efficiently on HPC clusters. Intel is working with the vendor and open source communities to resolve this issue and to lay the foundation for increasingly large neural networks acting on petabyte-scale datasets.

[3] For details, see https://www.intel.com/content/www/us/en/high-performance-computing-fabrics/omni-path-architecture-performance-overview.html

 

The post AI: Deeper Learning with Intel® Omni-Path Architecture appeared first on HPCwire.

Kathy Yelick Charts the Promise and Progress of Exascale Science

Fri, 09/15/2017 - 12:30

On Friday, Sept. 8, Kathy Yelick of Lawrence Berkeley National Laboratory and the University of California, Berkeley, delivered the keynote address on “Breakthrough Science at the Exascale” at the ACM Europe Conference in Barcelona. In conjunction with her presentation, Yelick agreed to a short Q&A discussion with HPCwire.

The timing of Yelick’s talk is timely as one year ago, on Sept. 7, 2016,  the U.S. Department of Energy made the first in a series of announcements about funding support for various components of the Exascale Computing Program, or ECP. The ECP was established to develop the exascale applications, system software, and hardware innovations necessary to enable the delivery of capable exascale systems.

Yelick is the Associate Laboratory Director for Computing Sciences, which includes the National Energy Research Scientific Computing Center (NERSC), the Energy Sciences Network (ESnet) and the Computational Research Division, which does research in applied mathematics, computer science, data science, and computational science. Yelick is also a professor of Electrical Engineering and Computer Sciences at the University of California at Berkeley. Her research is in parallel programming languages, compilers, algorithms and automatic performance tuning. Yelick was director of NERSC from 2008 to 2012. She was recently elected to the National Academy of Engineering (NAE) and the American Association of Arts and Sciences, and is an ACM Fellow and recipient of the ACM/IEEE Ken Kennedy and Athena awards.

HPCwire: What scientific applications necessitate the development of exascale?

Kathy Yelick: There are more than 20 ECP applications that, broadly speaking, fall into the areas of national security, energy, the environment, manufacturing, infrastructure, healthcare and scientific discovery. Associated with each is an exascale challenge problem—something that requires around 50 times the computational power of current systems. They include a diverse set of problems such as a 100-year simulation of the integrity of fields where petroleum is extracted or energy waste is stored; a predictive simulation of an urban area that includes buildings, water quality and electricity demands; and detailed simulations of the universe to better explain and interpret the latest observational data. There are also applications analyzing data at an unprecedented scale, from the newest light sources to complex environmental genomes, and cancer research data that includes patient genetics, tumor genomes, molecular simulations and clinical data.

These applications will help us develop cleaner energy, improve the resilience of our infrastructure, develop materials for extreme environments, adapt to changes in the water cycle, understand the origin of elements in the universe, and develop smaller, more powerful accelerators for use in medicine and industry. And as a California resident, I’m interested in the work to better assess the risks posed by earthquakes.

These projects are not simply scaling or porting old codes to new machines, but each of represents a new predictive or analytic capability. Several are completely new to high performance computing, and others add new capabilities to existing codes, integrating new physical models that are often at widely different space or time scales than the original code.

HPCwire: How do you respond to concerns that exascale programs are too focused on the hardware or will only benefit so-called hero codes?

Yelick: That’s an interesting statement in that ECP is currently committed to funding over $200 million this year to support applications development, software and hardware R&D in partnerships with vendors. There will be substantial machine acquisitions outside the project, but the project itself is directed at these other parts of the ecosystem. As I noted earlier, the application portfolio is not directed at a few hero codes, but represents a broad range of applications from both traditional and non-traditional HPC problem domains.

The NERSC facility is not slated to get one of the first exascale systems, but we expect to provide such a capability a few of years later with the NERSC-10 acquisition. Similarly, NSF is planning a leadership scale acquisition in roughly the same time frame, which should also benefit from the ECP investments. The investments made now in exascale R&D and software will benefit all exascale systems, and lessons learned on the initial applications will inform other teams. NERSC has experience going back to the introduction of massive parallelism in helping the community make such a transition and has already started preparing the user community through its NERSC Exascale Science Applications Program, NESAP. NESAP has 20 user code teams, some of which overlap with the ECP applications, partnered with NERSC and the vendors to prepare their codes for exascale.

HPCwire: What is your perspective on the progress that is being made toward exascale, given the challenges (power, concurrency, fault-tolerance, applications)?

Yelick: We are making great progress in our applications, which were the subject of a recent internal project review. Several of the application teams have found new levels of concurrency and memory optimizations to deal with the most recent DOE HPC system, the NERSC Cori machine with its 68-core nodes and high-bandwidth memory. Much of the ECP software and programming technology can be leveraged across multiple applications, both within ECP and beyond. For example, the Adaptive Mesh Refinement Co-Design Center (AMReX) which was launched last November is releasing its new framework to support the development of block-structured AMR algorithms at the end of September. At least five of the ECP application projects are using AMR, allowing them to efficiently simulate fine-resolution features.

Some of the R&D projects are also getting a better handle on the type of failures that will be important in practice. The hardware R&D on processor and memory designs have made great strides in reducing total system power, but it remains a challenge, and the resulting architecture innovations continue to raise software challenges for the rest of the team. Overall, we’re seeing the benefit of collaborations across the different parts of the project, incorporation of previous research results, and the need for even tighter integration across these parts.

HPCwire: There’s an expectation that exascale supercomputers will need to support simulation, big data and machine learning workloads, which currently have distinct software stacks. What are your thoughts on this challenge? Will container technology be helpful?

Yelick: Containers can certainly help support a variety of software stacks, including today’s analytics stack, and NERSC’s Shifter technology has helped bring this to its HPC systems. But I think we’ll also see new software developed for machine learning to achieve much higher performance levels and move them over to lighter-weight software. Porting Spark or TensorFlow to an exascale system will bring new user communities, but may not produce the most efficient use of these machines.

It’s somewhat ironic that training for deep learning probably has more similarity to the HPL benchmark than many of the simulations that are run today, although requirements for numerical precision are different and likely to lead to some architectural divergence. The algorithms in this space are evolving rapidly and projects like CAMERA (the Center for Advanced Mathematics for Energy Research Applications) are developing methods for analyzing data from some of the large DOE experimental facilities. Some of our policies around use of HPC need to change to better fit data workloads, both to handle on-demand computing for real-time data streams and to address the long-term needs for data provenance and sharing. The idea of receiving HPC allocations for a year at a time, and having jobs that sit in queues, will not work for these problems. NERSC is exploring all of these topics, such as with their recent 15-petaflop deep learning run described in a paper [and covered by HPCwire] by a team from NERSC, Intel and Stanford; a pilot for real-time job queues; automated metadata analysis through machine learning; and their NESAP for Data partnerships.

HPCwire: Speaking of machine learning and adapting codes to exascale, you’re the PI for the ECP applications project “Exascale Solutions for Microbiome Analysis,’ which also involves Los Alamos National Lab and DOE’s Joint Genome Institute. Can you tell us more about that project and how you’re tailoring Meraculous for exascale systems?

Yelick: The ExaBiome project is developing scalable methods for genome analysis, especially the analysis of microorganisms, which are central players in the environment, food production and human health. They occur naturally as “microbiomes,” cooperative communities of microbes, which means that sequencing an environmental sample produces a metagenome with thousands or even millions of individual species mixed together. Many of the species cannot be cultured in a lab and may never have been seen before—JGI researchers have even discovered new life forms from such analyses. To help understand the function of various genes, Aydin Buluc and Ariful Azad in the Computational Research Division have developed a new high performance clustering algorithm called HipMCL. Such bioinformatics analysis has often been viewed as requiring shared memory machines with large memory, but we have found that using clever parallel algorithms and HPC systems with low-latency interconnects and lightweight communication, we can scale these algorithms to run across petascale systems.

The algorithms are very different than most physical simulations because they involve graph walks, hash tables and highly unstructured sparse matrices. The de novo metagenome assembly challenge is to construct the individual genomes from the mixture of fragments produced by sequencers; it is based on an assembler called Meraculous, developed by Dan Rokhsar’s group at JGI and UC Berkeley. As part of the ExaBiome project we’ve built a scalable implementation extended to handle metagenomes called MetaHipMer (Metagenome High Performance Meraculous). These tools will enable the analysis of very complex environmental samples, and analysis over time, to understand how the microbial community changes with the rest of the environment and influences that environment.

The algorithms also reflect an important workload for future exascale machines. As described in our recent EuroPar 2017 paper, they require fine-grained communication and therefore can take advantage of high injection rates, low latency and remote atomic operations (e.g., remotely incrementing a counter) in the networks. The computation is entirely dominated by these operations and local string alignment algorithms, so there’s no floating point in the entire application. It’s important that we keep all of these workloads in mind as we push towards exasacle, to ensure the machines are capable of graph problems, bioinformatics and other highly irregular computational patterns that may be of interest outside of science and engineering communities.

HPCwire: What are some of the other key points from your talk that you’d like to share with our readers?

Yelick: First, the science breakthroughs from exascale programs will rely not just on faster machines, but also on the development of new application capabilities that build on prior research in mathematics, computer science and data science. We need to keep this research pipeline engaged over the next few years, so that we continue to have a vibrant research community to produce the critical methods and techniques that we will need to solve computational and data science challenges beyond exascale.

In that same vein, we shouldn’t think of exascale as an end goal, but rather as another point in the continuum of scientific computing. While much of DOE’s computing effort is currently devoted to exascale, we are already looking beyond to specialized digital architectures, quantum and neuromorphic computing, and new models of scientific investigation and collaboration for addressing future challenges.

The post Kathy Yelick Charts the Promise and Progress of Exascale Science appeared first on HPCwire.

Super Micro Computer, Inc. Receives Non-Compliance Letter From Nasdaq

Fri, 09/15/2017 - 06:50

SAN JOSE, Calif., Sept. 15, 2017 — Super Micro Computer, Inc. (NASDAQ:SMCI), a global leader in high-performance, high-efficiency server, storage technology and green computing, today announced that the Company received a notification letter from Nasdaq stating that the Company is not in compliance with Nasdaq listing rule 5250(c)(1), which requires timely filing of reports with the U.S. Securities and Exchange Commission. The September 14, 2017 letter was sent as a result of the Company’s delay in filing its Annual Report on Form 10-K for the period ending June 30, 2017 (the “Form 10-K”). The Form 10-K was due on August 29, 2017. The Company filed a Form 12b-25 on August 29, 2017, the effect of which was to extend the due date for the Form 10-K to September 13, 2017. The Company was unable to file the Form 10-K by September 13, 2017, for the reasons reported in the Form 12b-25 and further described below.

The Nasdaq notice has no immediate effect on the listing or trading of the Company’s common stock on the Nasdaq Global Select Market. Under the Nasdaq rules, the Company has 60 days from the date of the notice either to file the Form 10-K or to submit a plan to Nasdaq to regain compliance with Nasdaq’s listing rules. If a plan is submitted and accepted, the Company could be granted up to 180 days from the Form 10-K’s due date to regain compliance. If NASDAQ does not accept the Company’s plan, then the Company will have the opportunity to appeal that decision to a NASDAQ hearings panel.

As previously disclosed by the Company, additional time is needed for the Company to compile and analyze certain information and documentation and finalize its financial statements, as well as complete a related audit committee review, in order to permit the Company’s independent registered public accounting firm to complete its audit of the financial statements to be incorporated in the Form 10-K and complete its audit of the Company’s internal controls over financial reporting as of June 30, 2017. The Company is unable at this time to provide a date as to when the review and the audits will be completed.

There is no update to the Company’s previously announced guidance for net sales for the first quarter of fiscal year 2018 ending September 30, 2017. However, as a result of additional efforts required to finalize the financial statements and complete the audit committee review, the Company expects operating expenses for the quarter to increase due to higher legal and accounting costs.

The Company intends to file its 10-K promptly upon completion of the audit committee’s review and the completion of the audit.

 

About Super Micro Computer, Inc.

Supermicro, a global leader in high-performance, high-efficiency server technology and innovation is a premier provider of end-to-end green computing solutions for Data Center, Cloud Computing, Enterprise IT, Hadoop/Big Data, HPC and Embedded Systems worldwide. Supermicro’s advanced Server Building Block Solutions offer a vast array of components for building energy-efficient, application-optimized, computing solutions. Architecture innovations include Twin, TwinPro, FatTwin, Ultra Series, MicroCloud, MicroBlade, SuperBlade, Simply Double, Double-sided Storage, Battery Backup Power (BBP) modules and WIO/UIO. Products include servers, blades, GPU systems, workstations, motherboards, chassis, power supplies, storage, networking, server management software and SuperRack cabinets/accessories delivering unrivaled performance and value.

Source: Super Micro Computer

The post Super Micro Computer, Inc. Receives Non-Compliance Letter From Nasdaq appeared first on HPCwire.

Pages