Azure high-performance computing at Supercomputing 2020

Posted on November 16, 2020

Sr. Product Marketing Manager, Azure Marketing

Customers around the world rely on Microsoft Azure to drive innovations related to our environment, public health, energy sustainability, weather modeling, economic growth, and more. Finding solutions to these important challenges requires huge amounts of focused computing power. Customers are increasingly finding the best way to access such high-performance computing (HPC) is through the agility, scale, security, and leading edge performance of Azure’s purpose-built HPC and AI cloud services.

Azure’s market-leading vision for HPC and AI is based on a core of genuine and recognized HPC expertise, using proven HPC technology and design principles, enhanced with the best features of the cloud. The result is a capability that delivers performance, scale, and value unlike any other cloud. This means applications scaling 12 times higher than other public clouds. It means higher application performance per node. It means powering AI workloads for one customer with a supercomputer fit to be among the top five in the world. And it means delivering massive compute power into the hands of medical researchers over a weekend to prove out life-saving innovations in the fight against COVID-19.

Big moments for Azure HPC and AI Supercomputing in 2020

OpenAI

PlaceholderLast year, you might recall that Microsoft entered into a strategic partnership with OpenAI. Earlier this year, OpenAI released their first application programming interface (AP)I for their general-purpose language model, Generative Pre-trained Transformer 3 (GPT-3), giving developers access to advanced natural language processing (NLP) technologies for building more interactive applications and services. Shortly thereafter, OpenAI announced it has agreed to license its GPT-3 technology to Microsoft to use within its own products and services.

This will continue to offer developers and scientists new means to not just transform their organizations but to do so faster and with less disruption. It was also announced that Microsoft has deployed dedicated cloud supercomputing capability for OpenAI, comparable to the top five supercomputers in the world.

Read the blog here.

Duke University

PlceholderLiving with a pandemic causes people to be clever about creating new solutions. Ventilators have become a scarce resource as the COVID-19 pandemic spreads around the world. Professor Randles and her team at Duke University rushed to find a safe and easy way to effectively split a single ventilator between two patients. They partnered with Microsoft AI for Health to run over 800,000 computer hours in just 36 hours to build a life-saving solution.

Watch the video here.

COVID Moonshot

PlaceholderUCB transforms the lives of people living with severe diseases. They have been collaborating with biopharma, tech companies, chemical vendors, and individual contributors on the COVID Moonshot initiative, which is aimed at developing solutions to the COVID-19 pandemic. As soon as support and partnership with Microsoft was granted, the productivity of their teams went through the roof. By using Azure, they could now do in three days what would have taken six months. This computational horsepower combined with UCB Medicinal Chemistry efforts has helped move forward their research into their efficacy against COVID-19.

Watch the video here.

Previous 2020 Azure HPC and AI product launches

Azure HBv2 HPC clusters with over 80,000 AMD “Rome” cores and low latency InfiniBand interconnect.

Azure’s HBv2 HPC clusters have been a gamechanger for HPC customers. These true HPC cloud computing products mean the largest workloads can now run on the Azure cloud. Many firsts were achieved with HBv2, such as the first cloud to deliver over 80,000 cores for a single tightly coupled MPI application.

“In the world of Computational Fluid Dynamics (CFD), there is constant pressure to accelerate the speed of product design and today, our customers are looking turn around high-fidelity simulations in hours, not weeks. Running Simcenter STAR-CCM+ on the latest Microsoft Azure HBv2 series has enabled our customers to scale-up their simulations quickly and easily without expensive hardware investment or compromising solution fidelity. Our customers get the same performance and scaling as on-premises at lower cost, enabling them to make better engineering decisions faster.”—Keith Foston, Senior Product Manager, Simcenter STAR-CCM+, Siemens

Azure NVv4 Virtual Machines.

Azure NVvv4 Virtual Machines are an extreme choice in price and performance for your visualization needs.

"The diversity of the new AMD-based Worskpot cloud desktops on Microsoft Azure is a huge deal for us.  Based on the application requirements of each engineer, we can dedicate all or a fraction of the AMD GPU to their Workspot workstation on Azure.  This finer resolution of control gives us the financial edge we need to remove more people to Workspot cloud desktops on Azure and increase our overall productivity."Eric Quinn, CTO at C&S Companies

Azure NDv2 HPC and AI clusters with NVIDIA V100 Tensor Core GPUs.

Azure NDv2 GPU clusters are designed for the most demanding distributed HPC, AI, and machine learning workloads. Azure NDv2 GPU clusters with Riskfuel technology offer risk calculations 20 million times faster valuation of derivatives.

“The current market volatility demonstrates the need for real-time valuation and risk management for OTC derivatives. The era of the nightly batch is ending. And it’s not just the blazing fast inferencing of the Azure ND40rs_v2 Virtual Machine that we value so much, but also the model training tasks as well. On this fast GPU instance, we have reduced our training time from 48 hours to under four! The reduced time to train the model coupled with on-demand availability maximizes the productivity of our AI engineering team.”—Ryan Ferguson, CEO, Riskfuel

New Azure leadership unveiled at Supercomputing 2020 (SC20)

Delivering real AI supercomputing to customers with A100 Tensor Core GPUs and NVIDIA Mellanox HDR 200Gb/s InfiniBand networkingPlaceholder

Microsoft is proud to be among the first to deliver the power of NVIDIA’s A100 Tensor Core GPU to AI customers with the newly announced public preview of NDv4 A100 Virtual Machine.

Azure at the forefront of AMD EPYC “Milan” CPUs for HPC customers

PlaceholderMicrosoft and AMD are delighted to announce that Azure will be at the forefront of bringing AMD’s 3rd Generation EPYC “Milan” processor to HPC customers.

Azure HPC Collaboration Center Program

We are also announcing the launch of the Azure HPC Collaboration Center Program. The program will partner with industry-recognized experts to test the limits of the Azure HPC and AI Platform, share technical successes, document best practices, and showcase partner capabilities. Fostering a strong partner ecosystem is a core ethos of Microsoft’s approach to HPC and AI. The seed funding and close collaboration under this program is a key component of delivering these benefits to our HPC and AI customers and the wider community. Look out during SC20 week for announcements of the first Azure HPC Collaboration Centers.

Partner products and activities unveiled at SC20

Rescale

Azure HBv2 clusters for Compute Intensive workloads now generally available via Rescale.

Rescale eliminates complexity to enable organizations to accelerate science and engineering breakthroughs, with intelligent full-stack automation for big compute and the large scale power of cloud HPC. From supersonic jets to personalized medicine, industry leaders accelerate new product innovations with unprecedented speed and efficiency with the Rescale platform on Azure—a solution for intelligent full-stack automation for big compute and research and development collaboration. Azure's HBv2 clusters which are optimized for HPC workloads are now generally available on the Rescale platform both as on-demand and as spot instances. Rescale is also available through Azure Marketplace for seamless evaluation and purchase.

You can find more information here, or contact partners@rescale.com.

Mentor Graphics

Mentor Graphics launches SystemVision in Azure.

Mentor, A Siemens Business offers an advanced electronic circuit design and simulation environment that enables both engineers and electronic component manufacturers to collaborate on a level that was previously not possible. The company uses the Microsoft Azure platform to deploy SystemVision Cloud to users around the world, accessible via a web browser, with nothing to install. No longer limited to static images and PDF documents, electronic component manufacturers can now showcase their complex electronic components via live, interactive reference designs embedded on their own web properties. Everyone from electronic component experts to system architects can work together on circuit design and analysis from anywhere in the world on SystemVision.

“Microsoft Azure provided us a level of performance and platform support that enabled us to deliver SystemVision Cloud to our customers with a significantly reduced time to market."—Adam Cabler, Cloud Architect, Electronic Board Systems, Mentor A Siemens Business

Learn more about Azure HPC and AI from the experts

Join us at Supercomputing 2020 to learn more about Azure HPC and AI

We will feature speakers in SC20 virtual sessions, provide links to downloads, case studies, blogs, and additional Azure HPC videos. Please stop by our virtual booth to chat with us. We’d love to hear from you.

  • Azure HPC and AI: The reality—hear learnings and perspectives from a recognized leader of the HPC community, ex-Cray CTO Steve Scott, now Technical Fellow and Corporate VP at Microsoft.
  • Azure AI Supercomputing: See how powerful AI services on Azure are dramatically increasing the value of traditional HPC outcomes. Discover what “AI at Scale” means with Chris Berner from OpenAI, and learn how the new NDv4 A100 Azure GPU clusters can empower your innovations.
  • Azure High-Performance Computing: Azure HPC isn’t running HPC in the cloud—it is a cloud that has been purpose-built for HPC. Learn about record-breaking performance results on the leading technologies and modern environments. Get a full understanding of what the Azure HPC can offer you directly from the product experts themselves, including in-depth coverage of compute, storage, networking, services, and AI resources.
  • Using quantum-inspired algorithms to solve problems on today’s classical high-performance computing systems gives us a view of what may be possible in a future with scaled quantum computing. We are getting closer to that goal with a new project born out of Microsoft’s partnership with Pacific Northwest National Laboratory (PNNL) in the Quantum Science Center.

Additional links