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

Researchers at the Atos Quantum Laboratory have successfully modeled ‘quantum noise’ and as a result, simulation is more realistic than ever before, and is closer to fulfilling researchers’ requirements. "We are thrilled by the remarkable progress that the Atos Quantum program has delivered as of today," said Thierry Breton, Chairman and CEO of Atos.The post Atos Quantum Learning Machine can now simulate real Qubits appeared first on insideHPC.

Gilles Fourestey from EPFL gave this talk at the Swiss HPC Conference. "LENSTOOL is a gravitational lensing software that models mass distribution of galaxies and clusters. It is used to obtain sub-percent precision measurements of the total mass in galaxy clusters and constrain the dark matter self-interaction cross-section, a crucial ingredient to understanding its nature."The post Scratch to Supercomputers: Bottoms-up Build of Large-scale Computational Lensing Software appeared first on insideHPC.

Today DDN announced the opening of a new facility in Colorado Springs, Colorado, including a significant expansion of lab, testing and benchmarking facilities. The enhanced capabilities will enable DDN to accelerate development efforts and increase in-house capabilities to mimic customer applications and workflows. "Our Enterprise, AI, HPC and Cloud customers have always relied upon us to develop the world’s leading data storage solutions at-scale, and for our long-term focus and sustained investments in research, technology and innovation,” said Alex Bouzari, chief executive officer, chairman and co-founder of DDN. “We are excited to add our new Colorado Springs facility to the DDN R&D centers worldwide and to expand our team of very talented engineers and technologists who will continue to drive innovation for our customers in the years to come.”The post DDN Builds New Engineering Facility in Colorado focused on AI, Cloud, and Enterprise Data Challenges appeared first on insideHPC.

Alberto Madonaa gave this talk at the Swiss HPC Conference. "In this work we present an extension to the container runtime of Shifter that provides containerized applications with a mechanism to access GPU accelerators and specialized networking from the host system, effectively enabling performance portability of containers across HPC resources. The presented extension makes possible to rapidly deploy high-performance software on supercomputers from containerized applications that have been developed, built, and tested in non-HPC commodity hardware, e.g. the laptop or workstation of a researcher."The post Shifter – Docker Containers for HPC appeared first on insideHPC.

Altair software is now part of the Inspire Unlimited software-as-a-service offering available on the Azure cloud. "Unlike the HyperWorks Unlimited Appliance, where performance is based on the number of nodes, Inspire’s scale requirements are based on the number of simultaneous users; there could be 1,000 engineers working together at a time,” says Sam Mahalingam from Altair. “We felt that the HPC environment in Azure was architected to meet the type of back-end requirements we needed for Inspire.” Altair uses Microsoft Azure Virtual Machines, with NV instances powered by NVIDIA Tesla M60 GPUs.The post Altair Steps up to Azure Cloud with Inspire Unlimited appeared first on insideHPC.

Haodong Tang from Intel gave this talk at the 2018 Open Fabrics Workshop. "Efficient network messenger is critical for today’s scale-out storage systems. Ceph is one of the most popular distributed storage system providing a scalable and reliable object, block and file storage services. As the explosive growth of Big Data continues, there're strong demands leveraging Ceph build high performance & ultra-low latency storage solution in the cloud and bigdata environment. The traditional TCP/IP cannot satisfy this requirement, but Remote Direct Memory Access (RDMA) can."The post Accelerating Ceph with RDMA and NVMe-oF appeared first on insideHPC.

NASA researchers are using AI technologies to detect gravitational waves. The work is described in a new article in Physics Review D this month. "This article shows that we can automatically detect and group together noise anomalies in data from the LIGO detectors by using artificial intelligence algorithms based on neural networks that were already pre-trained to classify images of real-world objects," said research scientist, Eliu Huerta.The post Using Ai to detect Gravitational Waves with the Blue Waters Supercomputer appeared first on insideHPC.

Brian Barrett from Amazon gave this talk at the 2018 OpenFabrics Workshop. "As network performance becomes a larger bottleneck in application performance, AWS is investing in improving HPC network performance. Our initial investment focused on improving performance in open source MPI implementations, with positive results. Recently, however, we have pivoted to focusing on using libfabric to improve point to point performance."The post Amazon and Libfabric: A case study in flexible HPC Infrastructure appeared first on insideHPC.

"Supercomputing capability plays a key role in ECMWF’s success and in its ability to implement its strategic vision to 2025. In order to meet the required flexibility for future growth, ECMWF’s data centre is being relocated to Bologna, Italy and will be operational in 2020. The new Director of Computing will be responsible for delivering this challenging transition, fit-out and transformation, whilst ensuring that ECMWF’s computing capability continues to support some of the most critical scientific advances of our time."The post Job of the Week: Director of Computing at ECMWF appeared first on insideHPC.

The mission of the eScience Centre is to support research through the development and provision of advanced ICT infrastructure, services and expertise. We provide services and expertise in the areas of High Performance Computing, Cloud Services, Data Services and e-science. We support large research, educational and government institutions, as well as the business community. The […]The post Architects for Research Infrastructure appeared first on insideHPC.

DK Panda from Ohio State University gave this talk at the Swiss HPC Conference. "This talk will provide an overview of challenges in accelerating Hadoop, Spark, and Memcached on modern HPC clusters. An overview of RDMA-based designs for Hadoop (HDFS, MapReduce, RPC and HBase), Spark, Memcached, Swift, and Kafka using native RDMA support for InfiniBand and RoCE will be presented. Enhanced designs for these components to exploit NVM-based in-memory technology and parallel file systems (such as Lustre) will also be presented."The post Exploiting HPC Technologies for Accelerating Big Data Processing and Associated Deep Learning appeared first on insideHPC.

Today PRACE announced that Prof. Dr. Xiaoxiang Zhu, German Aerospace Center (DLR) and Technical University of Munich (TUM), Germany, is the winner of the 2018 PRACE Ada Lovelace Award for HPC for her outstanding contributions and impact on HPC in Europe. "Prof Zhu and her team (SiPEO) develop explorative algorithms to improve information retrieval from remote sensing data, in particular those from the current and next generation of Earth observation missions."The post Dr. Xiaoxiang Zhu wins 2018 PRACE Ada Lovelace Award for HPC appeared first on insideHPC.

Christoph Lameter from Jump Trading LLC gave this talk at the OpenFabrics Workshop. "Recently new types of memory have shown up like HBM (High Bandwidth Memory), Optane, 3DXpoint, NVDIMM, NVME and various "nonvolatile" types memory. This talk gives a brief rundown on what is available and gives some example on how the vendors enable the actual use of this memory in the operating system (f.e. DAX and filesystems) and then show how an application would make use of this memory. In particular then we will be looking at what considerations are important for the use of RDMA to those memory devices."The post New Types of Memory, their support in Linux, and how to use them via RDMA appeared first on insideHPC.

In this video from the 2018 GPU Technology Conference, Prof. Taisuke Boku from the University of Tsukuba & JCAHPC and Duncan Poole, President of OpenACC describe how OpenACC is accelerating science. "Dr. Boku and his team are working on OpenACC and XcalableACC for accelerated computing. XcalableACC is an extension of XMP for accelerated clusters using OpenACC. Programmers can develop an application to insert XMP and OpenACC directives into a sequential program."The post Developing Faster Algorithms with OpenACC and XcalableACC appeared first on insideHPC.

Dave Turek from IBM gave this talk at the Swiss HPC Conference. "There is a shift underway where HPC is beginning to be addressed with novel techniques and technologies including cognitive and analytic approaches to HPC problems and the arrival of the first quantum systems. This talk will showcase how IBM is merging cognitive, analytics, and quantum with classic simulation and modeling to create a new path for computational science."The post The Transformation of HPC: Simulation and Cognitive Methods in the Era of Big Data appeared first on insideHPC.

The 34th International Conference on Massive Storage Systems and Technologies (MSST 2018) has posted their Speaker Agenda. The event takes place May 14-16 in Santa Clara, California. "Join the discussion on webscale IT, and the demand on storage systems from IoT, healthcare, scientific research, and the continuing stream of smart applications (apps) for mobile devices."The post Agenda Posted for Mass Storage Conference in Santa Clara appeared first on insideHPC.

Today Intel announced top-tier OEM adoption of Intel’s field programmable gate array (FPGA) acceleration in their server lineup. This is the first major use of reprogrammable silicon chips to help speed up mainstream applications for the modern data center. "We are at the horizon of a new era of data center computing as Dell EMC and Fujitsu put the power and flexibility of Intel FPGAs in mainstream server products,” said Reynette Au, vice president of marketing for the Intel Programmable Solutions Group. “We’re enabling our customers and partners to create a rich set of high-performance solutions at scale by delivering the benefits of hardware performance, all in a software development environment.”The post Intel FPGAs Goes Mainstream for Enterprise Workloads appeared first on insideHPC.

Today RAID Incorporated updated its ARI-400 Series of storage solutions. The newest release features declustered RAID, which allows for mixing and matching of multiple disk capacities and greatly reduces rebuild times, a much welcomed technology as larger disks continue to be released into the market.The post ARI-400 Series Features Declustered RAID Technology appeared first on insideHPC.

"HPC organizations that utilize cloud service providers (AWS, Azure, Google, Oracle, etc.) in conjunction with Adaptive Computing’s NODUS Cloud Bursting Solution can significantly reduce their on-premise cluster sizes and costs by as much as 40-50 percent, and burst the rest of their HPC workload to the cloud, by implementing this hybrid approach."The post Reduce Costs with Adaptive Computing’s NODUS Cloud Bursting Solution appeared first on insideHPC.

Bill Magro from Intel gave this talk at the 2018 OpenFabrics Workshop. "Artificial Intelligence and High Performance Data Analytics workloads in the cloud are being fed by a deluge of data emanating from the Internet-connected population or people and things. This talk highlights the broadening role of OpenFabrics, in general, and the Open Fabrics Interface, in particular, to rise to the challenge of meeting the emerging requirements and become the software foundation for high-performance cloud fabrics."The post Intel’s Bill Magro Presents: Software Foundation for High-Performance Fabrics in the Cloud appeared first on insideHPC.

Today the ISC Group rolled out their new ISC High Performance Travel Grant Program to enable university students and young researchers to attend the conference in Germany. "Starting this year, ISC Group will select two individuals to receive the ISC Travel Grant. For recipients traveling from Europe or North Africa, the maximum funding is 1000 euros per person, and for the rest of the world, it is 1500 euros per person. ISC Group will also provide the grant recipients free registration for the entire conference, which takes place from June 24 – June 28 in Frankfurt, Germany."The post ISC 2018 Launches Travel Grant Program for Students appeared first on insideHPC.

Axel Koehler from NVIDIA gave this talk at the Switzerland HPC Conference. "The technology originally developed for HPC has enabled deep learning, and deep learning is enabling many usages in science. Deep learning is also helping deliver real-time results with models that used to take days or months to simulate. The presentation will give an overview about the latest hard- and software developments for HPC and Deep Learning from NVIDIA and will show some examples that Deep Learning can be combined with traditional large scale simulations."The post Video: The Convergence of HPC and Deep Learning appeared first on insideHPC.

Today Supermicro announced their new Rack Scale Design (RSD) 2.1 with pooled all-flash NVMe composable storage support for applications like high throughput ingest, HPC, data analytics, video streaming, CDN, and software-defined storage (SDS) environments. “Occupying just 1U of rack space, our all-flash NVMe storage systems support 32 hot-swap 2.5" NVMe SSDs for a half petabyte of high-performance storage with Supermicro RSD 2.1 that can be shared by 12 hosts simultaneously,” said Charles Liang, president and CEO of Supermicro. “With dynamically composable server nodes, efficient storage utilization, and independently upgradeable compute resources, our RSD 2.1 solutions with advanced NVMe pooled storage are perfect for efficient and flexible hyperscale datacenters. In fact, we have already deployed these 32-drive systems running Hadoop workloads for a major automobile company.”The post Supermicro RSD 2.1 Pools All-Flash NVMe Composable Storage appeared first on insideHPC.

Christian Kniep from Docker Inc gave this talk at the 2018 Swiss HPC Conference. "This talk will recap the history of and what constitutes Linux Containers, before laying out how the technology is employed by various engines and what problems these engines have to solve. Afterward Christian will elaborate on why the advent of standards for images and runtimes moved the discussion from building and distributing containers to orchestrating containerized applications at scale. In conclusion attendees will get an update on how containers foster the convergence of Big Data and HPC workloads and the state of native HPC containers."The post Video: State of Containers and the Convergence of HPC and BigData appeared first on insideHPC.

In this video, DK Panda from X-ScaleSolutions describes how the company optimizes and supports open source software for HPC. “At X-ScaleSolutions, we align cutting edge technology into a convergent trajectory so that business analytics with Big Data/Deep Learning can be solved with high-performance and scalability while exploiting the latest HPC and Cloud technologies. The vision of X-ScaleSolutions is to bring innovative and efficient end-to-end solutions, services, support, and training to our customers.”The post DK Panda Launches X-ScaleSolutions Startup for HPC Applications appeared first on insideHPC.

While FPGAs have been renowned for how hard they are to program, help may be on the way. Today Acceliz launched an App Store marketplace called AccelStore for FPGA applications. Currently, the marketplace hosts accelerator functions for data compression, search-and-replace, image encoding and security running on both Amazon Web Services (AWS) and OVH platforms and delivering performance improvements from 10x to more than 1000x compared to traditional pure software implementations. "Our customers are constantly looking for ways to boost their efficiency and improve service quality. FPGAs therefore represent a powerful tool for them,” said Nicolas Savides, VP of Strategic Partnerships at OVH. “AccelStore is an essential part of our plan for making FPGA acceleration a reality for our customers.”The post Accelize Launches App Store for FPGAs appeared first on insideHPC.

The 2018 Perth HPC Conference has announced their Call for Submissions. Hosted by the Pawsey Supercomputing Centre, The event takes place August 28-29 in Perth, Australia. "The second annual HPCAIAC Perth Conference will bring industry luminaries and leading subject matter experts together from all over the world to explore major and emerging domains and share in-depth insights on AI, Data Sciences, HPC, Deep/Machine Learning, Visualization and more."The post Call for Submissions: 2018 Perth HPC Conference appeared first on insideHPC.

"Python remains a single threaded environment with the global interpreter lock as the main bottleneck. Threads must wait for other threads to complete before starting to do their assigned work. The result of this model is that production code is produced that is too slow to be useful for large simulations."The post Python Can Do It appeared first on insideHPC.

The eighth international Women in HPC Workshop will be held June 28 at ISC 2018 in Frankfurt, Germany. This workshop provides leaders and managers in the HPC community with methods to improve diversity and also to provide early career women with an opportunity to develop their professional skills and profile. "Through collaboration and networking, WHPC strives to bring together women in HPC and technical computing while encouraging women to engage in outreach activities and improve the visibility of inspirational role models."The post Women in HPC Workshop Returns to ISC 2018 appeared first on insideHPC.

In this video from the GPU Technology Conference, Bob Fletcher from Verne Global discusses why more and more HPC & AI workloads are moving to the company's datacenters in Iceland. “Today’s computational environments are changing rapidly as more companies are looking to utilize HPC and intensive applications across an increasingly wide variety of industries. At Verne Global we have fully optimized our campus to meet the specific requirements of the international HPC community.”The post Iceland’s Verne Global Steps up to run HPC & AI Workloads in the Cloud appeared first on insideHPC.

Torsten Hoefler from (ETH) Zürich gave this talk at the 2018 Swiss HPC Conference. "Deep Neural Networks (DNNs) are becoming an important tool in modern computing applications. Accelerating their training is a major challenge and techniques range from distributed algorithms to low-level circuit design. In this talk, we describe the problem from a theoretical perspective, followed by approaches for its parallelization."The post Video: Demystifying Parallel and Distributed Deep Learning appeared first on insideHPC.

Today Energy Secretary Rick Perry announced the DOE has issued a Request for Proposal for 2-3 Exascale machines. "Called CORAL-2, this RFP is for up to $1.8 billion and is completely separate from the $320 million allocated for the Exascale Computing Project in the FY 2018 budget. Those funds are mostly focused at application development and software technology for an exascale software stack."The post Video: DOE Issues RFP for Exascale Supercomputers appeared first on insideHPC.

In this podcast, the Radio Free HPC team reviews the HPC highlights of the GPU Technology Conference. "From Rich's perspective, the key announcement centered around new NVIDIA DGX-2 supercomputer with the NVSwitch interconnect."The post Radio Free HPC Looks at HPC Highlights from the GPU Technology Conference appeared first on insideHPC.

LBNL Communications Manager Jon Bashor has announced his retirement after 27 years with the national lab system. "As communications manager, Jon has been key to the visibility of Berkeley Lab’s computing program, both through written articles and other material produced by Jon and his team and his community leadership—including several years of organizing the DOE booth at SC, the annual supercomputing conference,” said Associate Lab Director Kathy Yelick.The post Jon Bashor Retires After 27 Years of Service to National Labs appeared first on insideHPC.

In this video from the GPU Technology Conference, David Gonzolez from Ziff describes how Dell EMC powers AI solutions at his company. "ZIFF is unique in its approach to Ai. By focusing on empowering product visionaries and software engineers, ZIFF can help organizations fully unlock the insights and automation trapped within their unstructured data." To accelerate it's unstructured database technology, Ziff uses the Dell PowerEdge C4140, which allows them meet the demands of cognitive computing workloads with a dense, accelerator-optimized 1U server supporting 4 GPUs and superior thermal efficiency.The post How ZIFF Powers AI with Dell EMC Technologies appeared first on insideHPC.

"The goal of this hands-on workshop was to help participants optimize their application codes to exploit the different levels of parallelism and memory hierarchies in the Xeon Phi architecture,” said CSI computational scientist Meifeng Lin. “By the end of the hackathon, the participants had not only made their codes run more efficiently on Xeon Phi–based systems, but also learned about strategies that could be applied to other CPU-based systems to improve code performance.”The post Researchers Tune HPC Codes for Intel Xeon Phi at Brookhaven Hackathon appeared first on insideHPC.

In this video from the Supercomputing Frontiers Europe conference, Prof. Thomas Sterling describes how the National Strategic Computing Initiative will help keep the USA on the forefront of technology. Along the way he answers these questions: "Is there a race between China and the US?" and "What is Exascale and what does it mean for the value of computing?"The post Video: Thomas Sterling on the Current State of HPC appeared first on insideHPC.

The Ohio State University is seeking an HPC Security Engineer in our Job of the Week. "OSC provides high-performance computing services for university researchers and industrial clients. The HPC Systems Team delivers OSC’s production HPC systems, networking, and storage services through system design, procurement, deployment, security, and systems administration. This security engineer position supports the overall mission of the team and in particular duties related to systems security and implementation of security policy."The post Job of the Week: HPC Security Engineer at Ohio State University appeared first on insideHPC.

At the Ohio Supercomputer Center Statewide Users Group spring conference this week, OSC clients in fields spanning everything from astrophysics to linguistics gathered to share research highlights and hear updates about the center’s direction and role in supporting science across Ohio. "SUG is a great vehicle for us to not only communicate to our clients about what is going on from a policy perspective or hardware roadmaps and new services, but for us to hear back from the clients about what they are doing,” said Brian Guilfoos, HPC client services manager at OSC.“Our normal interaction with someone is very technical – ‘This is the thing I’m trying to do, what I’m having a problem with, etc.’ Here we get to take a broader view and look at the science, and it’s good for our staff to be reminded what is being done with our services and what we are enabling.”The post Ohio Supercomputing Center Hosts User Group Meeting appeared first on insideHPC.

Pointwise has released Version 18.1, a major update to its computational fluid dynamics (CFD) mesh generation software. With its origins in a multiyear development effort funded by the U.S. Air Force for overset grid assembly, this latest Pointwise release includes a suite of tools that provide increased automation and flexibility in mesh topology creation, mesh types and mesh quality assessment. "This release is the culmination and commercial manifestation of a two-year $1.2 million research and development effort funded by the U.S. Air Force Materiel Command, Arnold Engineering Development Complex,” said Nick Wyman, Pointwise’s director of applied research. “That effort in turn was an extension of a previous, two-year, Small Business Innovation Research (SBIR) Phase II award. While focused on overset grid assembly, the capabilities added to the software are applicable to virtually any type of mesh generation.”The post New Pointwise Release Adds Mesh Automation and Adaptability for CFD appeared first on insideHPC.

Today the OpenMP Architecture Review Board (ARB) announced the appointment of Barbara Chapman to its Board of Directors. "We are delighted to have Prof. Chapman join the OpenMP Board”, says Partha Tirumalai, chairman of the OpenMP Board of Directors. “Her decades of experience in high-performance computing and education will enhance the value OpenMP brings to users all over the world.”The post Barbara Chapman Joins Board of OpenMP ARB appeared first on insideHPC.

In this episode of Let's Talk Exascale, Scott Baden of LBNL describes the Pagoda Project, which seeks to develop a lightweight communication and global address space support for exascale applications. "What our project is addressing is how to keep the fixed cost as small as possible, so that cutting-edge irregular algorithms can efficiently move many small pieces of data efficiently."The post Let’s Talk Exascale: Developing Low Overhead Communication Libraries appeared first on insideHPC.

In this video from the GPU Technology Conference, Thierry Pellegrino describes how Dell EMC customers are applying HPC technologies to AI workloads. "I'll just mention one of our customers, AeroFarms, who use a lot of our technology in order to bring the value of IoT into an environment where you can do machine learning, deep learning, artificial intelligence, and automatically grow crops in an environment that you would never think it would be possible."The post Thierry Pellegrino on the Move to AI with HPC at Dell EMC appeared first on insideHPC.

In this video from the GPU Technology Conference, Les Lorenzo from Supermicro describes the company's new High Density NVMe Storage Server. "Supermicro’s latest storage server platform supporting the Next Generation Small Form Factor (NGSFF) media offers a new approach to enterprise all-flash storage applications, by providing all of the benefits of cost optimized M.2 flash drives with an increase in capacity and density and adding the hot-swap capabilities required for enterprise-class applications."The post Video: Supermicro Steps up with High Density NVMe Storage Server appeared first on insideHPC.

Over at the IBM blog, Jeff Welser writes that nearly 100 startups, venture capitalists, and industry thought leaders are gathering today at the first IBM Q Summit Silicon Valley event. "IBM Q Summit attendees are gathering to discuss what to expect over the next five years and what it means to be "quantum ready." The discussion will inevitably also center on the emerging role of the quantum developer and what that means for future application development."The post IBM brings in Startups to Accelerate Quantum Computing appeared first on insideHPC.

The National Supercomputing Centre (NSCC) Singapore and the Pawsey Supercomputing Centre have signed a Memorandum of Understanding (MOU) to collaborate in the fields of supercomputing, networking, data analytics, scientific software applications and visualization. "With our national petascale HPC platform, NSCC’s collaboration with international supercomputing centres such as the Pawsey, will benefit and bring together people and researchers with different skillsets and expertise to solve problems of a scale previously not possible,” said Professor Tan Tin Wee, NSCC’s Chief Executive."The post Pawsey Supercomputing Centre teams with NSCC Singapore to Accelerate Science appeared first on insideHPC.

In this video from the GPU Technology Conference, Peter Lilian from NVIDIA describes how the company works with Dell EMC to deliver extreme performance solutions for the Energy sector. "Dell EMC and NVIDIA have expanded their collaboration by signing a new strategic agreement to include joint product development of new products and solutions that address the burgeoning workload and datacenter requirements, with GPU-accelerated solutions for HPC, data analytics, and artificial intelligence."The post How NVIDIA Powers HPC for Energy at Dell EMC appeared first on insideHPC.

Burton J. Smith has passed away. An internationally recognized leader in HPC architecture, Smith was a co-founder of Tera Computing (later renamed Cray, Inc) and most recently a Microsoft technical fellow. Smith was 77. "Burton was an amazing intellect but more importantly, he was an amazing person. Every moment was important to him and he lived them all passionately. He touched many lives, made a profound impact in the world and will be missed by many,” said Todd Holmdahl, Smith’s colleague at Microsoft.The post HPC Pioneer Burton Smith passes away appeared first on insideHPC.

Today Mellanox announced that the company’s InfiniBand and Ethernet solutions have been chosen to accelerate the new NVIDIA DGX-2 artificial intelligence system. DGX-2 is the first 2 Petaflop system that combines sixteen GPUs and eight Mellanox ConnectX adapters, supporting both EDR InfiniBand and 100 gigabit Ethernet connectivity. The technological advantages of the Mellanox adapters with smart acceleration engines enable the highest performance for AI and Deep Learning applications.The post Mellanox powers NVIDIA DGX-2 AI Supercomputer appeared first on insideHPC.