Edit

Introduction to the Azure high-performance computing (HPC) scenario

  • Article

This scenario focuses on modernizing your machine learning, visualization, and rendering workloads in Azure at any scale with HPC + AI through the Cloud Adoption Framework.

Defining HPC

High-performance computing (HPC) on Azure is the complete set of compute, networking, and storage resources integrated with workload orchestration services for applications that provide advanced analytics, graphic-intensive visualizations, and scalable rendering.

Industries such as automotive, energy, health and life sciences, silicon, finance, and manufacturing typically use Azure HPC workloads to run complex simulations such as 3D modeling or mathematical tasks with their known and familiar tools/processes to perform the intensive tasks with thousands of compute hours in a matter of days.

While the computation of their data is being orchestrated, each industry customer can focus more on analyzing and creating value from the results to prioritize and achieve their business goals in a cost effective and timely manner.

HPC narrative

As customers from different industries have specific needs that require HPC investment, there's also the concern of sustainability. Compared to deploying huge on-premises infrastructure, having your HPC workloads in the cloud helps to use minimal computing resources only as and when is needed with flexible scalability to support sustainability goals.

HPC in Azure also offers competitive pricing and performance compared to on-premises options, robust global regulatory compliance, and next-generation machine-learning tools to drive smarter simulations and empower intelligent decision making.

When customers choose the relevant Azure HPC solution for their specialized computing tasks, there will also need to be considerations for following the Cloud Adoption Framework to prepare decision makers and central IT for successful cloud adoption. See below for common customer HPC journeys.

Common Customer HPC Journeys

  • Energy industry organizations such as oil and gas benefit from an up-scaled, end-to-end AI ecosystem that provides nurturing new ideas into safety operations and taking vast data repositories into AI solutions that promote operational advances such as predictive maintenance, simulation workloads, and automating completion reports to reduce data extraction times from months to hours.
  • Finance organizations use HPC to modernizing approaches to risk management, portfolio optimization, and implementing effective compliance and governance practices to protect consumer data. Computationally intensive workloads in Azure HPC have reduced costs and increased/instantaneous scalability for on-demand pricing model calculations.
  • Manufacturing organizations utilize HPC to fuel real-time product simulations that reduce time to market and improve product quality. Thousands of on-demand simulation jobs are executed to deliver faster solutions and more improved insights for following health and safety regulations.

Next step: Integrate HPC into your cloud adoption journey

Get started with the below list of guidance found at specific points throughout the cloud adoption journey to help you be successful in the cloud adoption scenario for your HPC environment.

For more information, refer to the landing zone accelerator for additional information.

Design guidelines are also available to provide guidance on creating landing zones for your HPC billing, identity and access management, network topology/connectivity, platform automation, management group and subscription organization, governance disciplines, and security disciplines.