Video: Exascale Deep Learning for Climate Analytics
Thorsten Kurth Josh Romero gave this talk at the GPU Technology Conference. “We’ll discuss how we scaled the training of a single deep learning model to 27,360 V100 GPUs (4,560 nodes) on the OLCF Summit HPC System using the high-productivity TensorFlow framework. This talk is targeted at deep learning practitioners who are interested in learning what optimizations are necessary for training their models efficiently at massive scale.” [READ MORE…]
Featured Stories
Agenda Posted for PASC19 in Zurich
PASC19 has posted its Full Conference Program. The event takes place June 12-14 in Zurich, Switzerland. “PASC19 is the sixth edition of the PASC Conference series, an international platform for the exchange of competences in scientific computing and computational science, with a strong focus on methods, tools, algorithms, application challenges, and novel techniques and usage of high performance computing.” [READ MORE…]
Barcelona Supercomputing Centre opens Arm Centre of Excellence for HPC
The Barcelona Supercomputing Center (BSC) and Arm Research have signed a multi-year agreement to establish the Arm-BSC Centre of Excellence. The move reflects a long history of BSC pioneering Arm technologies in HPC with the MontBlanc project. “The establishment of this Centre of Excellence will enable us to continue our close collaboration in order to advance the presence of Arm technologies in High Performance Computing and High Performance Data Analytics in Europe and worldwide.” [READ MORE…]
Mellanox Introduces High Speed Ethernet Cloud Fabric Technology
Today Mellanox introduced breakthrough Ethernet Cloud Fabric (ECF) technology based on Spectrum-2, the world’s most advanced 100/200/400 Gb/s Ethernet switches. ECF technology provides the ideal platform to quickly build and simply deploy state of the art public and private cloud data centers with improved efficiency and manageability. Spectrum-2 is at the heart a new family of SN3000 switches that come in leaf, spine, and super-spine form factors.” [READ MORE…]
Featured Resource
In-Network Computing and Next Generation HDR 200G InfiniBand
With the exponential growth of data that needs to be analyzed and the data resulting from ever-more complex workflows, the need for faster data movement has never been more challenging and critical to the worlds of High Performance Computing (HPC) and machine learning. Mellanox Technologies is once again moving the bar forward with the introduction of and end-to-end HDR 200G InfiniBand product portfolio. Download the new white paper, courtesy of Mellanox, that explores in-network computing and the benefits of the switch from 100G to 200G Infiniband.
Industry Perspectives
Epic 2018 HPC Road Trip begins at Idaho National Lab
In this special guest feature, Dan Olds from OrionX begins his Epic HPC Road Trip series with a stop at Idaho National Laboratory. “The fast approach of SC18 gave me an idea: why not drive from my home base in Beaverton, Oregon, to Dallas, Texas and stop at national labs along the way? I love a good road trip and what could be better than a 5,879 mile drive with visits to supercomputer users mixed in?” [READ MORE…]
Aerospace Moving Forward with Modeling and Simulation
In this special guest feature from Scientific Computing World, Gemma Church writes that the aerospace industry is stuck in the past – but it isn’t due to a lack of new simulation and modeling techniques. “While there have been incremental improvements, in terms of creating quieter and more fuel-efficient aeroplanes, there is not a lot of innovation coming in, compared to the last century when we created aeroplanes moving at the speed of sound and made air travel available to the masses.” [READ MORE…]
Featured from insideBIGDATA
- Data Scientists Will Get Very High Pay Around In These Top CitiesIn this contributed article, Stephanie Donahole a Business Analyst at Tatvasoft Australia, believes there has never been a better time to learn the concept of data science than today. The job landscape seems quite promising; opportunities span multiple industries and the nature of the job often allows for remote work flexibility and even self-employment. The […]
News from insideBIGDATA
- UC Berkeley Graduate Receives ACM Doctoral Dissertation Award
- LevaData Introduces Cognitive Sourcing for New Product Introduction Teams
- MarkLogic® Enhances Data Hub with Embedded Machine Learning for Smarter Data Integration
- Data Scientists Will Get Very High Pay Around In These Top Cities
- Unleash AI for Impact
- Take Your Business Use Cases to the Next Level with AI & ML
- How the EU is Mastering the Challenge of Trustworthy Artificial Intelligence
Editor’s Choice
HPE to Acquire Cray for $1.3 Billion
Today HPE announced that the company has entered into a definitive agreement to acquire Cray for approximately $1.3 billion. “This pending deal will bring together HPE, the global HPC market leader, and Cray, whose Shasta architecture is under contract to power America’s two fastest supercomputers in 2021,” said Steve Conway from Hyperion Research. “The Cray addition will boost HPE’s ability to pursue high-end procurements and will speed the combined company’s development of next-generation technologies that will benefit HPC and AI-machine learning customers at all price points.” [READ MORE…]
Achieving ExaOps with the CoMet Comparative Genomics Application
Wayne Joubert’s talk at the HPC User Forum described how researchers at the US Department of Energy’s Oak Ridge National Laboratory (ORNL) achieved a record throughput of 1.88 ExaOps on the CoMet algorithm. As the first science application to run at the exascale level, CoMet achieved this remarkable speed analyzing genomic data on the recently launched Summit supercomputer. [READ MORE…]
Video: Can FPGAs compete with GPUs?
John Romein from ASTRON gave this talk at the GPU Technology Conference. “We’ll discuss how FPGAs are changing as a result of new technology such as the Open CL high-level programming language, hard floating-point units, and tight integration with CPU cores. Traditionally energy-efficient FPGAs were considered notoriously difficult to program and unsuitable for complex HPC applications. We’ll compare the latest FPGAs to GPUs, examining the architecture, programming models, programming effort, performance, and energy efficiency by considering some real applications.” [READ MORE…]
EuroHPC – The EU Strategy in High Performance Computing
Thomas Skordas from the European Commission gave this talk at EuroHPC Summit Week in Poland. “The EuroHPC Summit Week brings together relevant European supercomputing stakeholders and decision makers, allowing them to share on the one hand their needs and future expectations and on the other hand the latest technological developments, to define synergies and participate in shaping the future of European supercomputing.” [READ MORE…]
Design Work Completed for SKA Telescope Supercomputer
An international group of scientists led by the University of Cambridge has finished designing the ‘brain’ of the Square Kilometre Array (SKA), the world’s largest radio telescope. When complete, the SKA will enable astronomers to monitor the sky in unprecedented detail and survey the entire sky much faster than any system currently in existence. “We estimate SDP’s total compute power to be around 250 PFlops – that’s 25% faster than IBM’s Summit, the current fastest supercomputer in the world.” [READ MORE…]
