High-Performance Computing News Analysis | insideHPC

Today, Intel announced that its AI technology is being used by Microsoft to power their new accelerated deep learning platform, called Project Brainwave. "Project Brainwave achieves a major leap forward in both performance and flexibility for cloud-based serving of deep learning models. We designed the system for real-time AI, which means the system processes requests as fast as it receives them, with ultra-low latency. Real-time AI is becoming increasingly important as cloud infrastructures process live data streams, whether they be search queries, videos, sensor streams, or interactions with users."The post Intel FPGAs Accelerate Microsoft’s Project Brainwave appeared first on insideHPC.

Today ANSYS 18.2 was released. This latest release brings increased levels of accuracy, speed and ease-of-use – spurring more engineers to use simulation across every stage of the product lifecycle to design cutting-edge products more efficiently and economically. "Our customers rely on ANSYS engineering simulation technology to cut costs, limit late-stage design changes, and tame the toughest engineering challenges. This latest release continues to build upon the industry's most accurate simulation portfolio, offering enhanced speed and accuracy – enabling more users, no matter their level of experience, to reduce development time and increase product quality."The post ANSYS 18.2 Enhances Simulation Speed and Accuracy appeared first on insideHPC.

Today NCSA and Hyperion Research released a new study that examines HPC and Industry partnerships. Aimed at identifying and understanding best practices in partnerships between public high performance computing centers and private industry, the study aims to promote the vital transfer of scientific knowledge to industry and the important transfer of industrial experience to the scientific community.The post New Study Uncovers Best Practices for Effective Partnerships between Public HPC Centers and Industrial Users appeared first on insideHPC.

Today Atos in Europe announced a new reseller agreement in which Dell EMC will resell Atos’ high-end 8 to 16 sockets x86 Bullion servers. "Over the last few years, Dell EMC and Atos have been working together to combine Bullion servers and Dell EMC unified storage solutions to provide our customers with a leading solution for deployment of mission-critical SAP HANA projects," said Ravi Pendekanti, Senior Vice President at Dell EMC.The post Dell EMC Joins Atos Global Reseller Alliance appeared first on insideHPC.

Over at the SC17 Blog, Brian Ban begins his series of SC17 Session Previews with a look at a talk on High Performance Big Data. "Deep learning, using GPU clusters, is a clear example but many Machine Learning algorithms also need iteration, and HPC communication and optimizations."The post SC17 Invited Talk Preview: High Performance Machine Learning appeared first on insideHPC.

"Like the previous disruptions of clusters vs. monolithic systems or Linux vs. proprietary operating systems, cloud changes the status quo, takes us out of our comfort zone, and gives us a sense of lack of control. But the effect of price, the flexibility to dynamically change your system size and choose the best architecture for the job, the availability of applications, the ability to select system cost based on the needs of a particular workload, and the ability to provision and run immediately, will prove very attractive for HPC users."The post Gabriel Broner on Why Cloud is the Next Disruption in HPC appeared first on insideHPC.

In this video, Piush Patel describes how Intel Xeon Scalable Processors speed up Altair RADIOSS structural simulations with Hyperworks. "RADIOSS is a leading structural analysis solver for highly non-linear problems under dynamic loadings. It is used across all industries worldwide to improve the crashworthiness, safety, and manufacturability of structural designs."The post Intel Xeon Scalable Processors Speed Altair RADIOSS with Hyperworks appeared first on insideHPC.

Frameworks, applications, libraries and toolkits—journeying through the world of deep learning can be daunting. If you’re trying to decide whether or not to begin a machine or deep learning project, there are several points that should first be considered. This is the second article in a five-part series that covers the steps to take before launching a machine learning startup. This article covers popular machine learning technology.The post Machine Learning Technology: A Guide to Scaling Up and Out appeared first on insideHPC.

Despite the cloud hype, legacy HPC apps are alive and well. While it may seem like they can’t mix, the process of bursting these applications to the cloud is bringing these staples to cloud table. Avoiding rewrites can efficiently bring immediate HPC cloud benefits to organizations big and small. "Many technologies and solutions now available allow for the functional and highly efficient coordination and connection between legacy applications and the well-known advantages of the cloud."The post Unlike Oil and Water, Legacy and Cloud Mix Well appeared first on insideHPC.

In this video, from Andrew Gibiansky from Baidu describes baidu-allreduce, a newly released C library that enables faster training of neural network models across many GPUs. The library demonstrates the allreduce algorithm, which you can embed into any MPI-enabled application. ""Neural networks have grown in scale over the past several years, and training can require a massive amount of data and computational resources. To provide the required amount of compute power, we scale models to dozens of GPUs using a technique common in high-performance computing but underused in deep learning."The post Video: Baidu Releases Fast Allreduce Library for Deep Learning appeared first on insideHPC.

Fayetteville State University is using Bright Cluster Manager for ground-breaking Artificial Intelligence research. Bright Cluster Manager enables the University to deploy complete clusters over bare metal and manage them effectively. Managing the hardware, operating system, HPC software, and its users is done with an elegant easy-to-use graphical user interface. With Bright Cluster Manager, IT/system administrators can quickly get clusters up and running and keep them running reliably throughout their lifecycle.The post Fayetteville State University Leverages Bright Cluster Manager for AI Research appeared first on insideHPC.

PRACE is offering an online Supercomputing 101 course through the Future Learn program. This free online course will introduce you to what supercomputers are, how they are used and how we can exploit their full computational potential to make scientific breakthroughs. "Using supercomputers, we can now conduct virtual experiments that are impossible in the real world."The post PRACE Offers Supercomputing 101 Course appeared first on insideHPC.

In this podcast, the Radio Free HPC team discusses the eclipse and how researchers are using supercomputers to simulate the corona of the sun at eclipse time. After that, look at the the top Technology stories in our Catch of the Week:The post Radio Free HPC Looks at Eclipse Simulations appeared first on insideHPC.

DownUnder GeoSolutions (America) is seeking a Senior HPC Administrator in our Job of the Week. "If you are passionate about leading edge technology you will love this role. What you will receive, along with a fantastic group of colleagues who will support and encourage you is the chance to be working on one of the largest private clusters in the world!"The post Job of the Week: Senior HPC Administrator at DownUnder GeoSolutions appeared first on insideHPC.

Geoffrey Hinton from the University of Toronto gave this talk at the Vector Institute. "What is Wrong with 'standard' Convolutional Neural Nets? They have too few levels of structure: Neurons, Layers, and Whole Nets. We need to group neurons in each layer in 'capsules' that do a lot of internal computation and then output a compact result."The post Video: What is Wrong with Convolutional Neural Nets? appeared first on insideHPC.

"Meeting the Enterprise Challenges of HPC System developers and users face obstacles deploying complex new HPC technologies, such as: energy efficiency, reliability and resiliency requirements, or developing software to exploit HPC hardware. All can delay HPC adoption. But the HPC Alliance will help. HPE and Intel will collaborate with you on workstream integration, solution sizing and software-to-hardware integration. We will help, whether you seek ease of everything, to mix workloads, or have fast-growing, increasingly complex systems."The post Video: Customers Leverage HPE & Intel Alliance for HPC appeared first on insideHPC.

The New Zealand Science Infrastructure (NeSI) is commissioning a new HPC system that will be colocated at two facilities. "The new systems, provide a step change in power to NeSI’s existing services, including a Cray XC50 Supercomputer and a Cray CS400 cluster High Performance Computer, both sharing the same high performance and offline storage systems."The post NeSI in New Zealand Installs Pair of Cray Supercomputers appeared first on insideHPC.

Predictive Sciences ran a large-scale simulation of the Sun’s surface in preparation for a prediction of what the solar corona will look like during the eclipse. "The Solar eclipse allows us to see levels of the solar corona not possible even with the most powerful telescopes and spacecraft,” said Niall Gaffney, a former Hubble scientist and director of Data Intensive Computing at the Texas Advanced Computing Center. “It also gives high performance computing researchers who model high energy plasmas the unique ability to test our understanding of magnetohydrodynamics at a scale and environment not possible anywhere else.”The post Researchers Use TACC, SDSC and NASA Supercomputers to Forecast Corona of the Sun appeared first on insideHPC.

In this video, you'll learn how to start submitting deep neural network (DNN) training jobs in Azure by using Azure Batch to schedule the jobs to your GPU compute clusters. "Previously, few people had access to the computing power for these scenarios. With Azure Batch, that power is available to you when you need it."The post Video: Deep Learning on Azure with GPUs appeared first on insideHPC.

"Now available as Open Source, PBS Pro software optimizes job scheduling and workload management in high-performance computing environments – clusters, clouds, and supercomputers – improving system efficiency and people’s productivity. Built by HPC people for HPC people, PBS Pro is fast, scalable, secure, and resilient, and supports all modern infrastructure, middleware, and applications."The post RCE Podcast Looks at PBS Professional Job Scheduler appeared first on insideHPC.

SC17 will feature a panel discussion entitled How Serious Are We About the Convergence Between HPC and Big Data? "The possible convergence between the third and fourth paradigms confronts the scientific community with both a daunting challenge and a unique opportunity. The challenge resides in the requirement to support both heterogeneous workloads with the same hardware architecture. The opportunity lies in creating a common software stack to accommodate the requirements of scientific simulations and big data applications productively while maximizing performance and throughput.The post SC17 Panel Preview: How Serious Are We About the Convergence Between HPC and Big Data? appeared first on insideHPC.

Today SIGHPC announced that Ilkay Altintas is the inaugural winner of the ACM SIGHPC Emerging Woman Leader in Technical Computing Award. "I am thrilled that Dr. Altintas has been selected to receive the inaugural SIGHPC Emerging Woman Leader in Technical Computing Award,” commented Dr. Jeffrey Hollingsworth, interim CIO at the University of Maryland and Chair of SIGHPC, in reaction to the award committee’s decision. “Dr. Altintas' work enables technical collaboration among different groups; this is critical not only for high performance computing, but science in general. With leaders like Dr. Altintas, the future of our field is in good hands."The post Dr. Ilkay Altintas to recieve ACM SIGHPC Emerging Woman Leader in Technical Computing Award appeared first on insideHPC.

Oak Ridge National Laboratory is moving equipment into a new high-performance computing center this month which is anticipated to become one of the world’s premier resources for open science computing. "There were a lot considerations to be had when designing the facilities for Summit,” explained George Wellborn, Heery Project Architect. “We are essentially harnessing a small city’s worth of power into one room. We had to ensure the confined space was adaptable for the power and cooling that is needed to run this next generation supercomputer.”The post ORNL Readies Facility for 200 Petaflop Summit Supercomputer appeared first on insideHPC.

Santiago Nuñez-Corrales, a graduate researcher at NCSA and Ph.D. student in Informatics at the University of Illinois at Urbana-Champaign was one of twelve recipients to be awarded the SIGHPC/Intel Computational and Data Science Fellowship for 2017. "The fellowship brings with it so many benefits," said Nuñez-Corrales. "Attending SC17 is important to keep up with development in the HPC community, and the financial support is critical to continuing my degree, as I have no funding from my home country."The post Nuñez-Corrales Awarded SIGHPC Computational Data and Science Fellowship appeared first on insideHPC.

In this video, researchers describe how the new HPC facility at Rockefeller University will power bioinformatics research and more. This is the first time that Rockefeller University has purpose-built a datacenter for high performance computing.The post Video: HPC Powers Bioinformatics Research at Rockefeller University appeared first on insideHPC.

Researchers at RENCI are using xDCI Data CyberInfrastructure to manage brain microscopy images that were overwhelming the storage capacity at individual workstations. “BRAIN-I is a computational infrastructure for handling these huge images combined with a discovery environment where scientists can run applications and do their analysis,” explained Mike Conway, a senior data science researcher at RENCI. “BRAIN-I deals with big data and computation in a user-friendly way so scientists can concentrate on their science.”The post xDCI Infrastructure Manages 3D Brain Microscopy Images at RENCI appeared first on insideHPC.

Kathy Yelick from LBNL will give the HPC keynote on Exascale computing at the upcoming ACM Europe Conference. With main themes centering on Cybersecurity and High Performance Computing, the event takes place Sept. 7-8 in Barcelona.The post Kathy Yelick to Keynote ACM Europe Conference appeared first on insideHPC.

Employing a hybrid of MPI across nodes in a cluster, multithreading with OpenMP* on each node, and vectorization of loops within each thread results in multiple performance gains. In fact, most application codes will run slower on the latest supercomputers if they run purely sequentially. This means that adding multithreading and vectorization to applications is now essential for running efficiently on the latest architectures.The post More Than Ever, Vectorization and Multithreading are Essential for Performance appeared first on insideHPC.

A new computing resource is available for Purdue researchers running applications that can take advantage of GPU accelerators. The system, known as Halstead-GPU, is a newly GPU-equipped portion of Halstead, Purdue’s newest community cluster research supercomputer. Halstead-GPU nodes consist of two 10-core Intel Xeon E5 CPUs per node, 256 GB of RAM, EDR Infiniband interconnects and two NVIDIA Tesla P100 GPUs. The GPU nodes have the same high-speed scratch storage as the main Halstead cluster.The post Purdue Adds New Resource for GPU-accelerated Research Computing appeared first on insideHPC.

In this video, Scott Harrison from Rock Flow Dynamics and Bruno Franzini from NICE Software explain how they scale HPC workloads in the cloud. "You'll learn how they leverage Amazon EC2 Spot instances and Amazon S3 to create cost-effective, scalable clusters that power tNavigator, Rock Flow Dynamics' solution for running dynamic reservoir simulations. You'll also see how they use NICE Software's DCV to stream the OpenGL-based user interface to interact with 3D models."The post Video: HPC Reservoir Simulations on AWS with NICE appeared first on insideHPC.

Illinois Rocstar is seeking a Sr. Scientific Programmer in our Job of the Week. "Have you ever thought that you’d like to write code to model something awesome? Like rockets? Aircraft? Explosions? Advanced battery performance? That’s what we do at Illinois Rocstar! Producing tools, interfaces, and systems to run on machines ranging from high-end engineering workstations to the most advanced high performance computing platforms in the world, we are all about making High Performance Computational Science and Engineering better and easier to use."The post Job of the Week: Sr. Scientific Programmer at Illinois Rocstar appeared first on insideHPC.

Professor Philip Diamond, Director General of the international Square Kilometer Array (SKA) project, will be the keynote speaker at SC17. "Professor Diamond, accompanied by Dr. Rosie Bolton, SKA Regional Centre Project Scientist, will take SC17 attendees around the globe and out into the deepest reaches of the observable universe as they describe the SKA’s international partnership that will map and study the entire sky in greater detail than ever before."The post Philip Diamond from SKA Project to Keynote SC17 appeared first on insideHPC.

Today Dell EMC announced it will build a supercomputer to power Swinburne University of Technology’s groundbreaking research into astrophysics and gravitational waves. "We will be looking for gravitational waves that help us learn more about supernovas, the formation of stars, intergalactic gases and more,” said Professor Bailes. “It’s exciting to think that we as OzGRav could make the next landmark discovery in gravitational wave astrophysics – and the Dell EMC supercomputer will allow us to capture, visualise and process the data to make those discoveries.”The post Dell EMC Supercomputer to Power OzGRav Studies of Black Holes appeared first on insideHPC.

In this video from Evan Schneider at Princeton University, hot galactic winds destroy a cool cloud of interstellar gas. “The process of generating galactic winds is something that requires exquisite resolution over a large volume to understand—much better resolution than other cosmological simulations that model populations of galaxies,” Robertson said. “This is something you really need a machine like Titan to do.”The post Video: Gas Clouds Wither in Galactic Winds appeared first on insideHPC.

Researchers are using XSEDE supercomputers to better understand the forces at work at the center of the Milky Way galaxy. The work could reveal how instabilities develop in extreme energy releases from black holes. "While nothing – not even light – can escape a black hole’s interior, the jets somehow manage to draw their energy from the black hole."The post How Extreme Energy Jets Escape a Black Hole appeared first on insideHPC.

Today Microsoft announced it has acquired Cycle Computing, a software company focused on making cloud computing resources more readily available for HPC workloads. "Now supporting InfiniBand and accelerated GPU computing, Microsoft Azure looks to be a perfect home for Cycle Computing, which started its journey with software for aggregating compute resources at AWS. The company later added similar capabilities for Azure and Google Cloud."The post Microsoft Acquires Cycle Computing appeared first on insideHPC.

In order to better keep a finger on the pulse of the Earth’s health, NASA developed DeepSat, a deep learning AI framework for satellite image classification and segmentation. DeepSat provides vital signs of changing landscapes at the highest possible resolution, enabling scientists to use the data for independent modeling efforts.The post DeepSat: Monitoring the Earth’s Vitals with AI appeared first on insideHPC.

Frameworks, applications, libraries and toolkits—journeying through the world of deep learning can be daunting. If you’re trying to decide whether or not to begin a machine or deep learning project, there are several points that should first be considered. This is the first article in a five-part series that covers the steps to take before launching a machine learning startup.The post What Developers Need to Consider When Exploring Machine Learning appeared first on insideHPC.

Today Insilico Medicine announced a research collaboration with The Bitfury Group, the world’s leading full-service blockchain technology conglomerate, to develop novel solutions for healthcare applications. The companies signed a memorandum of understanding (MOU) to collaborate in the academic and commercial settings to develop AI on Blockchain solutions for the healthcare industry.Blockchain can secure and streamline our medical systems, while AI has the potential to revitalize data management and machine learning to help identify trends and diseases,” said Valery Vavilov, founder, and CEO of The Bitfury Group. “By partnering with Insilico, we will be able to combine their expertise in deep learning and bioinformatics with our Blockchain proficiency and real-time solutions to create bespoke and innovative new products for the healthcare sector.”The post AI and Blockchain Companies Partner to Advance Healthcare Research appeared first on insideHPC.

In this Texas Standard podcast, Dan Stanzione from TACC describes Stampede2, the most powerful university supercomputer in the United States. "Phase 1 of the Stampede2 rollout, now complete, features 4,200 Knights Landing (KNL) nodes, the second generation of processors based on Intel's Many Integrated Core (MIC) architecture. Later this summer Phase 2 will add 1,736 Intel Xeon Skylake nodes."The post Podcast: 18 Petaflop Stampede 2 Supercomputer Powers Research at TACC appeared first on insideHPC.

A broad array of system administrators, developers, researchers and students who share an interest in the MVAPICH open-source library for high performance computing will gather this week for the fifth meeting of the MVAPICH Users Group (MUG). "Dr. Panda’s library is a cornerstone for HPC machines around the world, including OSC’s systems and many of the Top 500,” said Dave Hudak, Ph.D., interim executive director of OSC. “We’ve gained a lot of insight and expertise from partnering with DK and his research group throughout the years.”The post OSC Hosts fifth MVAPICH Users Group appeared first on insideHPC.

In this podcast, the Radio Free HPC team looks alarming new hacks of health care data. With news that Biohackers have successfully inserted malware into DNA, security is becoming a matter of concern for everything from scanners to gene sequencers. After that, we do our Catch of the Week.The post Radio Free HPC Looks at Biohacking appeared first on insideHPC.

"The Intel HPC Developer Conference 2017 has become the place for developers to learn about the latest trends, developments, and the future of HPC. By submitting an abstract and speaking at the conference you will be able to reach top-level decision makers within the HPC industry."The post Call for Abstracts: Intel HPC Developer Conference at SC17 appeared first on insideHPC.

In this video from the Heroes of Deep Learning series, Andrew Ng interviews Head of Baidu Research, Yuanqing Lin. "Baidu Research brings together global research talent to work on AI technologies in areas such as image recognition, speech recognition, high performance computing, natural language processing and deep learning. Baidu Research comprises four labs: the Silicon Valley AI Lab, the Institute of Deep Learning, the Big Data Lab and the Augmented Reality Lab."The post Heroes of Deep Learning: Yuanqing Lin appeared first on insideHPC.

"Traveling to Mars and further destinations will require more sophisticated computing capabilities to cut down on communication latencies and ensure astronauts’ survival, but existing computing resources are limited and incapable of extended periods of uptime. Settled aboard the SpaceX Dragon Spacecraft, the Spaceborne Computer is a year-long experiment from HPE and NASA—roughly the amount of time it will take to get to Mars—which will test a supercomputer’s ability to function in the harsh conditions of space."The post Video: HPE goes for Mission to Mars with Supercomputer Launch appeared first on insideHPC.

This week Slovenia became the ninth member state to sign the European Declaration on High Performance Computing. Designed to coordinate HPC efforts throughout Europe, the EuroHPC declaration was originally launched during Digital Day in March 2017 by France, Germany, Italy, Luxembourg, the Netherlands, Portugal and Spain. Belgium signed the declaration in June. "Providing only 5% of global HPC resources, but consuming about 30% of them, EU needs to to step up cooperation in HPC to boost Europe's scientific capabilities and industrial competitiveness. The EuroHPC goal is to to develop a high-performance computing ecosystem based on European technology, including low power chips and to have exascale supercomputers based on European technology in the global top 3 by 2022."The post Slovenia signs European Declaration on HPC appeared first on insideHPC.

Today Mellanox announced the availability of storage reference platforms based on its revolutionary BlueField System-on-Chip (SoC), combining a programmable multicore CPU, networking, storage, security, and virtualization acceleration engines into a single, highly integrated device. "BlueField is the most highly integrated NVMe over Fabrics solution,” said Michael Kagan, CTO of Mellanox. “By tightly integrating high-speed networking, programmable ARM cores, PCIe switching, cache, memory management, and smart offload technology all in one chip; the result is improved performance, power consumption, and affordability for flash storage arrays. BlueField is a key part of our Ethernet Storage Fabric solution, which is the most efficient way to network and share high-performance storage.”The post Mellanox BlueField Accelerates NVMe over Fabrics appeared first on insideHPC.

In this video, David Henty from EPCC conducts a video-based quiz on MPI. "The multiple-choice questions are partly designed for fun to test attendees' knowledge, but are mainly aimed at promoting discussion about MPI and its usage in real applications. All that is assumed is a working knowledge of basic MPI functionality: send, receive, collectives, derived datatypes and non-blocking communications."The post Test Your Knowledge with the MPI Quiz appeared first on insideHPC.

The DOE has awarded 1 Billion CPU hours of compute time on Oak Ridge supercomputers to a set important research projects vital to our nation's future. ALCC allocations for 2017 continue in the tradition of innovation and discovery with projects awards ranging from 2 million to 300 million processor hours.The post DOE Awards 1 Billion Hours of Supercomputer Time for Research appeared first on insideHPC.

The Ohio Supercomputer Center played a critical role in helping researchers reach a milestone mapping the growth of the universe from its infancy to present day. "The new results released Aug. 3 confirm the surprisingly simple but puzzling theory that the present universe is composed of only 4 percent ordinary matter, 26 percent mysterious dark matter, and the remaining 70 percent in the form of mysterious dark energy, which causes the accelerating expansion of the universe."The post OSC Helps Map the Invisible Universe appeared first on insideHPC.