Inside HPC & AI News | High-Performance Computing & Artificial Intelligence

In this special guest feature from Scientific Computing World, Laurence Horrocks-Barlow from OCF predicts that containerization, cloud, and GPU-based workloads are all going to dominate the HPC environment in 2020. "Over the last year, we’ve seen a strong shift towards the use of cloud in HPC, particularly in the case of storage. Many research institutions are working towards a ‘cloud first’ policy, looking for cost savings in using the cloud rather than expanding their data centres with overheads, such as cooling, data and cluster management and certification requirements."The post Predictions for HPC in 2020 appeared first on insideHPC.

GigaIO has developed a new whitepaper to describe GigaIO FabreX, a fundamentally new network architecture that integrates computing, storage, and other communication I/O into a single-system cluster network, using industry standard PCIe (peripheral component interconnect express) technology.The post The GigaIO FabreX Network – New Frontiers in Networking For Big Data appeared first on insideHPC.

The ISC Workshop On In Situ Visualization 2020 has issued its Call for Participation. The event takes place June 25 in Frankfurt, Germany. "We encourage contributed talks on methods and workflows that have been used for large-scale parallel visualization, with a particular focus on the in situ case. Presentations on codes that closely couple numerical methods and visualization are particularly welcome."The post Call for Participation: ISC Workshop on In Situ Visualization 2020 appeared first on insideHPC.

Richard S. Sutton from DeepMind Alberta gave this talk NeurIPS 2019. "In practice, I work primarily in reinforcement learning as an approach to artificial intelligence. I am exploring ways to represent a broad range of human knowledge in an empirical form--that is, in a form directly in terms of experience--and in ways of reducing the dependence on manual encoding of world state and knowledge."The post Video: Toward a General AI-Agent Architecture appeared first on insideHPC.

Ken Raffenetti from Argonne gave this talk at ATPESC 2019. "The Argonne Training Program on Extreme-Scale Computing (ATPESC) provides intensive, two-week training on the key skills, approaches, and tools to design, implement, and execute computational science and engineering applications on current high-end computing systems and the leadership-class computing systems of the future."The post Video: Overview of HPC Interconnects appeared first on insideHPC.

CIRC at Washington State University is seeking an HPC Systems Administrator in our Job of the Week. "Ideal candidates should have in-depth experience with the provisioning and administration of HPC clusters. Applicants who have experience with CentOS or RHEL, high speed networking using Mellanox Infiniband, resource schedulers such as Slurm, automation tools such as SaltStack, and parallel file systems including BeeGFS and Spectrum Scale, are highly encouraged to apply."The post Job of the Week: HPC Systems Administrator at Washington State University appeared first on insideHPC.

Fostering STEM education is a key to the future of high performance computing. Along these lines, ISC 2020 HPC Career Day will give 200 job seekers interested in HPC the opportunity to participate in the ISC 2020 conference on Wednesday, June 24. "Organizations scouting for STEM talent are welcome to utilize this new program to make connections with students, early and mid-career professionals looking for exciting prospects within the areas of HPC, machine learning, and data analytics."The post ISC 2020 Launches HPC Career Day appeared first on insideHPC.

Purdue University’s CERIAS Center for Education and Research in Information Assurance and Security has announced the addition of a new laboratory facility that dramatically increases Purdue’s cyber-physical research, emulation, and analysis capabilities. "This new laboratory is a mirror of the facilities already within Sandia National Labs that have served as the platform for joint CERIAS and DOE research since 2017,” said Theresa Mayer, executive vice president for research and partnerships at Purdue University. “The opening of SOL4CE at Purdue allows us to increase both the speed and impact of our national security research collaboration with Sandia National Labs.”The post Purdue University to open Scalable Open Laboratory for Cyber Experimentation appeared first on insideHPC.

A partnership XSEDE and the Institute for Research on Innovation and Science (IRIS) will examine how access to advanced research computing resources and services available via XSEDE affect the collaboration networks and scientific productivity of participating researchers. "IRIS will link the IRIS UMETRICS dataset containing transaction-level administrative data on sponsored research projects from dozens of the nation’s leading higher educational institutions to data from XSEDE allocations. This will result in a new way to examine how access to supercomputers influences the way researchers collaborate with colleagues and the productivity of individuals and research teams."The post IRIS and XSEDE to investigate the impact of research supercomputing appeared first on insideHPC.

Researchers at Argonne National Laboratory have developed a new molecular layer etching technique that could potentially enable the manufacture of increasingly small microelectronics. "Our ability to control matter at the nanoscale is limited by the kinds of tools we have to add or remove thin layers of material. Molecular layer etching (MLE) is a tool to allow manufacturers and researchers to precisely control the way thin materials, at microscopic and nanoscales, are removed,” said lead author Matthias Young, an assistant professor at the University of Missouri and former postdoctoral researcher at Argonne.The post New Argonne etching technique could advance semiconductors appeared first on insideHPC.

In this video from SC19, Berkeley researchers visualizes an entire brain at nanoscale resolution. The work was published in the journal, Science. "At the core of the work is the combination of expansion microscopy and lattice light-sheet microscopy (ExLLSM) to capture large super-resolution image volumes of neural circuits using high-speed, nano-scale molecular microscopy."The post Visualizing an Entire Brain at Nanoscale Resolution appeared first on insideHPC.

In this special guest feature, Jeff Reser from SUSE describes how Linux and HPC are key enabling technologies behind the research and breakthroughs in Genomics. "The Human Genome Project is an excellent example of large-scale international cooperation. It took a closely-coordinated and collaborative team effort to complete. Once the human genome had been successfully sequenced and decoded, it was immediately made publicly available. Since then, new information has been regularly published and made freely available. Here at SUSE, we’re totally committed to this community-driven “open source” ideal. It permeates everything we do."The post How HPC is Powering the Age of Genomic Big Data appeared first on insideHPC.

The US Department of Energy’s Exascale Computing Project (ECP) has announced the following staff changes within the Software Technology group. Lois Curfman McInnes from Argonne will replace Jonathan Carter as Deputy Director for Software Technology. Meanwhile Sherry Li is now team lead for Math Libraries. "We are fortunate to have such an incredibly seasoned, knowledgeable, and respected staff to help us lead the ECP efforts in bringing the nation’s first exascale computing software environment to fruition,” said Mike Heroux from Sandia National Labs.The post Exascale Computing Project Announces Staff Changes Within Software Technology Group appeared first on insideHPC.

The U.S. Department of Energy’s Office of Science, under the leadership of Under Secretary of Energy Paul Dabbar, sponsored around 70 representatives from multiple government agencies and universities at the first Quantum Internet Blueprint Workshop, held in New York City Feb. 5-6. The primary goal of the workshop was to begin laying the groundwork for a nationwide entangled quantum Internet.The post DOE Workshop Begins Mapping the Future of Quantum Communications appeared first on insideHPC.

Researchers at SDSC and the Wisconsin IceCube Particle Astrophysics Center have successfully completed a second computational experiment using thousands of GPUs across Amazon Web Services, Microsoft Azure, and the Google Cloud Platform. "We drew several key conclusions from this second demonstration,” said SDSC’s Sfiligoi. “We showed that the cloudburst run can actually be sustained during an entire workday instead of just one or two hours, and have moreover measured the cost of using only the two most cost-effective cloud instances for each cloud provider.”The post Second GPU Cloudburst Experiment Paves the Way for Large-scale Cloud Computing appeared first on insideHPC.

Today Award-winning Japanese HPC Cloud company XTREME-D announced an agreement with San Francisco-based Digital Realty, a leading global provider of data center, colocation, and interconnection solutions. MC Digital Realty will supply 10kW racks from its Digital Osaka 2 facility to host XTREME-D’s flagship product, XTREME-Stargate. XTREME-D has also formed a strategic alliance with Lenovo Enterprise Solutions Ltd. to be a supply partner for the product.The post XTREME-D Launches New HPC Infrastructure Services with Digital Realty appeared first on insideHPC.

Powered by storage technology supplied by Microway, the Northeast Storage Exchange (NESE) is changing the way Boston-area universities approach research data storage. "Born out of a groundbreaking regional high-performance computing project, NESE aims to break further ground—to create a long-term, growing, self-sustaining data storage facility serving both regional researchers and national and international-scale science and engineering projects."The post Microway powers Shared Research Computing Storage Project in Massachusetts appeared first on insideHPC.

"The paper addresses the inherent limitations associated with today's most popular gradient-based methods, such as Adaptive Moment Estimation (ADAM) and Stochastic Gradient Descent (SGD), which incorporate backpropagation. MemComputing's approach instead aims towards a more global and parallelized optimization algorithm, achievable through its entirely new computing architecture."The post Whitepaper: Accelerate Training of Deep Neural Networks with MemComputing appeared first on insideHPC.

A research team of academics from the University of Leeds and University of Nottingham believes its has found a way of delivering ultra- fast modulation, by combining the power of acoustic and light waves. The findings were published in Nature Communications. “This result opens a new area for physics and engineering to come together in the exploration of the interaction of terahertz sound and light waves, which could have real technological applications.”The post Using sound and light for ultra-fast data transfer appeared first on insideHPC.

The twelfth international Women in HPC workshop has issued its Call for Posters. The half-day WHPC workshop takes place June 25 at ISC20 in Frankfurt, Germany. "We are encouraging women who consider themselves to be ‘early career’ to participate, however this opportunity is open to help everyone who feels they may benefit from presenting their work, irrespective of career stage."The post Call for Posters: Women in HPC Workshop at ISC20 appeared first on insideHPC.

Colin Sauze from Aberystwyth University gave this talk at FOSDEM 2020. "The motivation for this was to overcome key problems faced by new HPC users. The talk will also discuss some of the technical challenges in deploying an HPC environment to a Raspberry Pi and attempts to keep that environment as close to a “real” HPC as possible. Methods to automate the installation process will also be covered."The post Introducing HPC with a Raspberry Pi Cluster appeared first on insideHPC.

Today OnScale announced their sponsorship of Revolution in Simulation, a collaborative community helping to increase the value of engineering simulation software investments through the democratization of simulation. OnScale is providing their expertise in CAE and funding to support the initiative. OnScale will participate as a moderator in the HPC topic as well as lead a new SME topical section coming soon at Rev-Sim.org on MultiPhysics Simulation.The post OnScale joins Revolution in Simulation community appeared first on insideHPC.

To use quantum computers on a large scale, we need to improve the technology at their heart – qubits. Qubits are the quantum version of conventional computers’ most basic form of information, bits. The DOE’s Office of Science is supporting research into developing the ingredients and recipes to build these challenging qubits.The post Stepping up Qubit research at the DOE appeared first on insideHPC.

Twenty-two postdoctoral fellows from across the Computing Sciences Area shared the status of their current projects at the first CSA Postdoc Symposium, held January 30-31 at Berkeley Lab. Their presentations covered a broad range of research topics, including code optimization, machine/deep learning, network routing, modeling and simulation of complex scientific problems, exascale, and other next-generation computer architectures.The post Postdoc Symposium at Berkeley Lab Looks to Exascale for Modeling and Simulation appeared first on insideHPC.

In this Let's Talk Exascale Podcast, Stuart Slattery and Damien Lebrun-Grandie from ORNL describe how they are readying algorithms for next-generation supercomputers at the Department of Energy. "The mathematical library development portfolio of the Software Technology (ST) research focus area of the ECP provides general tools to implement complex algorithms. These algorithms are designed to scale up for supercomputers so that ECP teams can then use them to accelerate the development and improve the performance of science applications on DOE high-performance computing architectures."The post Podcast: Solving Multiphysics Problems at the Exascale Computing Project appeared first on insideHPC.

Europe has developed a strategy for exascale computing, through partnerships and collaboration of European HPC vendors, academic institutions and HPC centers. It aims to deliver exascale-class systems and place the continent in the top three powers for supercomputing and science and industry using HPC. "It is a major step forward for Europe to reach the next level of computing capacity; it will help us to advance in future-oriented technologies, like the Internet of Things (Iot), AI, robotics and data analytics."The post Exascale in Europe appeared first on insideHPC.

The University of Florida has chosen Qumulo’s distributed file system for its scalable capacity, real-time data analytics, and industry-recognized commitment to customer care. "We wanted a storage solution that would not only work for UF’s faculty but work for our research computing staff as well," said Erik Deumens, scientist and director, UF Information Technology – Research Computing. "Selecting Qumulo takes the guesswork out of our storage management and makes us more efficient when scheduling diagnostic operations. We find the Qumulo system to be rich in features and easy to work with. It is a very cost-effective solution so that we are making the best use of university funds."The post University of Florida Accelerates BioTech Research with Qumulo appeared first on insideHPC.

Today PASC20 announced that this year’s Public Lecture will be presented by Dalia Conde, Professor at the University of Southern Denmark and Director of Science at Species360. The lecture will focus on her team’s efforts in fighting one of the greatest current concerns of our global community: biodiversity loss. "In this keynote talk, we will unveil the results of a global initiative aiming to map, quantify and disseminate species open information to conservation policymakers globally. By developing partnerships to map information and generate development platforms, workflows and storage between open biodiversity repositories, we will outline how computational methods can be applied to novel scientific domains."The post PASC20 talk to explore Data Landscapes to Rescue Species from Extinction appeared first on insideHPC.

Félix-Antoine Fortin from Compute Canada gave this talk at FOSDEM 2020. "Magic Castle uses the open-source software Terraform and HashiCorp Language (HCL) to define the virtual machines, volumes, and networks that are required to replicate a virtual HPC infrastructure. In this talk, we will give a live demonstration of the creation of a cluster."The post Compute Canada’s Magic Castle: Terraforming the Cloud for HPC appeared first on insideHPC.

The Royal Society Publishing has recently released a special compendium of articles based on a recent scientific discussion meeting with HPC Industry thought leaders. "This issue contains contributions from those who develop and implement numerical algorithms and software libraries – numerical analysts, computer scientists, and high-performance computing researchers - with those who use them in some of today's most challenging applications."The post Compendium of articles published on Numerical Algorithms for HPC Science appeared first on insideHPC.

A novel program at Stanford is finding a second life for used HPC clusters, providing much-needed computational resources for research while giving undergraduate students a chance to learn valuable career skills. To learn more, we caught up with Dellarontay Readus from the Stanford High Performance Computing Center (HPCC).The post Stanford Student Program Gives Supercomputers a Second Life appeared first on insideHPC.

The iRODS Consortium has welcomed KU Leuven as its newest Consortium member. As one of Europe’s oldest universities, Belgium’s KU Leuven boasts a long tradition of ground-breaking research and high quality education. With a new Research Data Management program, the university aims to accelerate science and innovation by enabling scientists to more effectively and efficiently find, access, share, and reuse data. "With a strong history of supporting the management and reuse of scientific data at universities in Europe and around the world, iRODS is well positioned to provide a framework for KU Leuven’s ambitious research data management initiative,” said Jason Coposky, Executive Director, iRODS Consortium. “We are pleased to welcome KU Leuven to the iRODS Consortium and look forward to working with them to guide the development of future capabilities that will benefit their researchers and the broader user community.”The post KU Leuven in Belgium joins iRODS Consortium appeared first on insideHPC.

Felix Abecassis and Jonathan Calmels from NVIDIA gave this talk at FOSDEM 2020. "We will present the challenges in doing distributed deep learning training at scale on shared heterogeneous infrastructure. At NVIDIA, we use containers extensively in our GPU clusters for both HPC and deep learning applications. We love containers for how they simplify software packaging and enable reproducibility without sacrificing performance."The post Distributed HPC Applications with Unprivileged Containers appeared first on insideHPC.

In this TACC podcast, researchers describe how they are using XSEDE supercomputers to run some of the highest resolution simulations ever of galaxy clusters. One really cool thing about simulations is that we know what's going on everywhere inside the simulated box," Butsky said. "We can make some synthetic observations and compare them to what we actually see in absorption spectra and then connect the dots and match the spectra that's observed and try to understand what's really going on in this simulated box."The post Podcast: Simulating Galaxy Clusters with XSEDE Supercomputers appeared first on insideHPC.

D.E. Shaw Research is seeking Systems Administrators for Servers, Clusters and Supercomputers in our Job of the Week. "Our research effort is aimed at achieving major scientific advances in the field of biochemistry and fundamentally transforming the process of drug discovery. Exceptional sysadmins sought to manage systems, storage, and network infrastructure for a New York–based interdisciplinary research group."The post Job of the Week: Systems Administrator at D.E. Shaw Research appeared first on insideHPC.

Immersive cooling vendor GRC has announced a new business collaboration with Avaso, an IT solutions provider with global service delivery and warranty capabilities. Through this collaboration, Avaso will provide installation and maintenance services, as well as offer extended warranty coverage for GRC's ICEraQ, ICEtank, HASHtank, and HASHraQ liquid immersion-cooling systems to customers in 150+ countries. In addition to servicing GRC’s systems, the service extends to the OEM immersed servers as well, enabling a single point of contact for maintenance and warranty related events.The post Avaso to Deliver Worldwide Support for GRC Immersive Cooling appeared first on insideHPC.

Studies have found that men tend to drastically outnumber women at technology conferences, limiting female viewpoints, minimizing women’s contributions to leadership, and exhibiting fewer female role models. At CENIC’s 2020 Conference: The Right Connection, more than one-third of the speakers will be female. "Diversity is good for technology, and CENIC is committed to supporting programs that increase diversity, inclusion, and equity for underrepresented populations in the field of computer networking."The post CENIC 2020 Conference to showcase Growing Contributions of Women in Computer Networking appeared first on insideHPC.

Today Sylabs announced the release of SingularityPRO 3.5, a popular container platform for HPC, supercomputing, and AI. "SingularityPRO 3.5, released January 21st, 2020, brings exciting new features to the long-term professionally supported version of the container platform. Based on the open source 3.5.2 release, SingularityPRO will receive security and bug fixes for 3 years, making it an ideal solution for the business-driven needs of enterprise customers containerizing their compute workloads."The post Sylabs releases SingularityPRO 3.5 appeared first on insideHPC.

In this video, Exascale Computing Project Director Doug Kothe describes how disciplined and tailored project management led to very impressive results in what was likely the most comprehensive independent review of the project to date. "ECP’s products will be robust, production ready, and functional right out of the box; and ECP is driving the sharing of information through regular training not only with ECP participants but also the broader US high-performance computing community to lower barriers to using exascale systems and accelerated architectures in general."The post Interview: Exascale Computing Project Update for 2020 appeared first on insideHPC.

In this podcast, the Radio Free HPC looks at the new ECMWF supercomputer. "This new system will give them roughly 5x more compute power than their current system. The new box is an Atos BullSequana XH2000 fueled by high-end AMD 7742 Epyc processors, which will be the most powerful weather computer in the world. During the conversation we look at the history of ECMWF vendors, discuss the implications on weather forecasts given the power of this new system and the computational difficulties inherent in weather prediction."The post Podcast: AMD to Power New ECMWF Supercomputer for Weather Forecasting appeared first on insideHPC.

Today Eni dedicated its new HPC5 system, the most powerful industrial supercomputer in the world. "HPC5 by Dell Technologies is made up of 1,820 Dell EMC PowerEdge C4140 servers, each with two Intel Gold 6252 24-core processors and four NVIDIA V100 GPU accelerators. The servers are connected through an InfiniBand Mellanox HDR ultra-high-performance network with a speed of 200 Gbit/s and a full non-blocking topology that ensures efficient and direct connection among every server. HPC5 also comes with a high-performance 15-petabyte storage system (200 GB/s aggregate read/write speeds)."The post Eni unveils HPC5 Supercomputer from Dell Technologies appeared first on insideHPC.

Today the Big Compute Conference announced sponsors and speakers for its inaugural event, held February 11-12, 2020 in San Francisco. The two-day conference will feature business leaders and scientists describing how they are transforming their industries with access to unlimited cloud compute. "Big Compute 20 brings together thought leaders in aerospace, automotive, AI, biotech, medical, academic, technology, and chemical industries. In addition to inspiring talks, the event will feature workshops, networking, panels and a hackathon sprint, all focused on the freedom to think big."The post Big Compute 20 Conference Announces Speaker Lineup appeared first on insideHPC.

An international team has completed the most comprehensive study of whole cancer genomes to date, significantly improving our fundamental understanding of cancer and signposting new directions for its diagnosis and treatment. The Barcelona Supercomputing Centre has been involved from the initial stages of this project, and has contributed with the analysis of data, with the design of specific computing solutions for cancer genomics, as well as in the answering of specific questions related to with the biology of tumors.The post BSC Powers Pan-Cancer Project appeared first on insideHPC.

"As researchers seek scalable, high performance methods for storing data, Ceph is a powerful technology that needs to be at the top of their list. Ceph is an open-source software-defined storage platform. While it’s not often in the spotlight, it’s working hard behind the scenes, playing a crucial role in enabling ambitious, world-renowned projects such as CERN’s particle physics research, Immunity Bio’s cancer research, The Human Brain Project, MeerKat radio telescope, and more."The post How Ceph powers exciting research with Open Source appeared first on insideHPC.

Researchers from Rice University have introduced a data-driven framework that formulates extreme weather prediction as a pattern recognition problem, employing state-of-the-art deep learning techniques. "In this paper, we show that with deep learning you can do analog forecasting with very complicated weather data — there's a lot of promise in this approach."The post Deep Learning for Predicting Severe Weather appeared first on insideHPC.

Today Atos announced the completion of its acquisition of Maven Wave, a U.S.-based cloud and technology consulting firm specialized in delivering digital transformation solutions for large enterprises. With this acquisition, Atos reinforces its global leadership in cloud-solutions for applications, data analytics and machine learning in hybrid and multi-cloud platforms. "Together, Maven Wave and Atos create the strongest Google Cloud services portfolio offered anywhere, providing customers proven expertise and knowledge in executing their digital transformation and delivering outstanding experiences to their customers,” commented Maven Wave Founders Brian Farrar, Jason Lee and Jeff Lee.The post Atos completes acquisition of Maven Wave appeared first on insideHPC.

Today Samsung Electronics launched 'Flashbolt,' its third-generation High Bandwidth Memory 2E (HBM2E). The new 16-gigabyte (GB) HBM2E is uniquely suited to maximize HPC systems and help system manufacturers to advance their supercomputers, AI-driven data analytics and state-of-the-art graphics systems in a timely manner. "With the introduction of the highest performing DRAM available today, we are taking a critical step to enhance our role as the leading innovator in the fast-growing premium memory market," said Cheol Choi, executive vice president of Memory Sales & Marketing at Samsung Electronics.The post Samsung Launches Flashbolt High Bandwidth 2E Memory appeared first on insideHPC.

Terrell Russell from iRODS gave this talk at SC19. "The Integrated Rule-Oriented Data System (iRODS) is open source data management software used by research organizations and government agencies worldwide. iRODS is released as a production-level distribution aimed at deployment in mission critical environments. It virtualizes data storage resources, so users can take control of their data, regardless of where and on what device the data is stored."The post Beyond Discoverability: Metadata to Drive Your Data Management appeared first on insideHPC.

Arm's HPC User Group (A-HUG) is transitioning to a fully-fledged community-led organization to better support the Arm ecosystem. The first annual meeting of the community-lead A-HUG will be held March 12-13 in Porto, Portugal. "The A-HUG event will include a hands-on training event and excellent talks covering the broad landscape of early systems & experience, near-term expectations for new hardware, and long term trends for architectures."The post Arm HPC User Group to Host First Annual Meeting in Portugal appeared first on insideHPC.

"Researchers, scientists, and developers are advancing science by accelerating their high performance computing applications on NVIDIA GPUs using specialized libraries, directives, and language-based programming models. From computational science to AI, CUDA-X HPC, OpenACC, and CUDA are GPU-accelerating applications to deliver groundbreaking scientific discoveries. And popular languages like C, C++, Fortran, and Python are being used to develop, optimize, and deploy these applications."The post How NVIDIA Enables Scientific Research for HPC Developers appeared first on insideHPC.