UberCloud and SUSE:  Next Generation HPC Accelerates Innovation

Wolfgang Gentzsch and Daniel Gruber, UberCloud, and Terry Smith, SUSE

HPC is an essential capability to enterprises across a broad spectrum of industries, including manufacturing, automotive, aerospace, energy, life sciences, finance, and more.  HPC gives enterprises the ability to discover scientific and business insights, develop new products, and open new avenues for progress by enabling multi-physics simulations, digital twin models, big data analytics, machine learning, and other complex engineering and scientific workloads.  But even the largest enterprises can struggle to deploy, manage, and scale the robust HPC resources they need for success in the modern world.

UberCloud’s next generation HPC engineering platform

UberCloud and SUSE deliver the next generation of enterprise HPC … a cloud-native approach … that empowers engineers and scientists to accelerate innovation, enables IT with streamlined tooling and enhanced service delivery, and provides organizations with unprecedented visibility to create efficiencies through a unified IT strategy.

UberCloud delivers an integrated solution of cloud-native software and services that help enterprises:

• Remove resource limitations
  Traditional HPC is difficult and costly to scale, manage, and keep up-to-date.  This forces engineers and scientists to limit simulation models to fit within available resources, reducing the accuracy of results and lowering the quality of insights for next-generation products.
  
  UberCloud and SUSE make it easy and safe to take advantage of the scale and power available in any public cloud (e.g., Amazon Web Services, Google Cloud, or Microsoft Azure) to enable engineers with the HPC compute and storage resources they need.  And, enterprises run the UberCloud Engineering Simulation Platform within their own cloud platform accounts, thus remaining fully and exclusively in control.

• Streamline management and administration
  HPC environments are notoriously complex, requiring specialized skills to support HPC hardware, software, and workflows.  Introducing the cloud to address HPC resource limitations only exacerbates these challenges, especially as IT must deal with structures and controls that are often unique to each cloud platform.
  
  The UberCloud Engineering Simulation Platform streamlines administration, regardless of the underlying public or private cloud, simplifying IT operations and even providing self-service capabilities that can empower engineers and scientists to quickly and easily leverage HPC resources within governed limits.

• Unify IT strategies
  HPC hardware, software, and operations can be quite outside the traditional enterprise IT experience, often resulting in the formation of IT silos and making it difficult to integrate HPC into an overall IT strategy that can address inefficiencies and make effective use of the cloud.
  
  The UberCloud Engineering Simulation Platform streamlines IT operations and also natively integrates with SUSE Rancher to provide unprecedented, multi-cloud observability and management.  Enterprises can gain holistic governance of their IT landscapes, enabling them to quickly adapt to changing conditions and driving efficiencies for lower total cost of ownership.
• Accelerate adoption
  New technologies can take time to learn, slowing adoption and limiting the benefits that can come through modernization.
  
  The UberCloud Engineering Simulation Platform easily integrates HPC cloud into existing IT operations and workflows – including the use of SUSE Rancher for multi-cloud governance.  A self-service user interface further reduces the burden on IT while empowering users to easily access company cloud resources.
  
  Unique UberCloud HPC application containers give scientists and engineers familiar HPC cluster environments, applications, and workflows, enabling them to continue their work with little or no additional training.  UberCloud HPC containers are built on SUSE Linux Enterprise base container images to deliver optimum performance, compatibility, and security.

UberCloud’s HPC engagement model

UberCloud leverages a proven on-boarding process with services delivered by HPC, cloud, and container experts focused on empowering enterprise customers to accelerate innovation.

UberCloud engagements feature:

• Joint discovery workshop
  In an initial meeting, UberCloud explores together with enterprise stakeholders their current challenges, needs, applications, and goals.
• Proof of concept
  Through a service engagement, UberCloud deploys a test environment, providing enterprise stakeholders an opportunity to try the solution.
• Production engineering
  After a successful proof of concept, UberCloud implements a customized, production environment within the customer’s own cloud accounts.
• Software and support
  UberCloud software and support subscriptions provide cost-efficient access to the UberCloud Engineering Simulation Platform, UberCloud HPC application containers, and support services.

UberCloud and SUSE recognize the importance of HPC to accelerate innovation across a broad spectrum of industries and endeavors - from manufacturing to energy production to life sciences to finance and beyond. With the UberCloud Engineering Simulation Platform, featuring SUSE Linux Enterprise and SUSE Rancher, enterprises can accelerate innovation with improved access to HPC resources, streamlined operations, and realized efficiencies lowering total cost of ownership.

Next steps for next generation HPC

• Learn how UberCloud and SUSE help 3DT Holdings revolutionize heart-valve surgery for better patient outcomes.
• Read how UberCloud enables Rimac Technology to optimize HPC in the cloud for developing the fastest electric hypercars in the world.
• Connect with UberCloud.
  
  

Author Bios

Wolfgang Gentzsch is President and Co-Founder of UberCloud. He was the Chairman of the International ISC Cloud Conference series. Previously, he was an Advisor to EU projects, Director of the German D-Grid Initiative and the North Carolina Statewide Grid, Member of the Board of Directors of the Open Grid Forum, Member of the US President's Council of Advisors for Science and Technology, and a professor of computer science and math at several universities in North Carolina and in Regensburg, Germany.

Daniel Gruber is Director of Architecture at UberCloud. He holds a B.Sc. and M.Sc. in Information Engineering. Daniel is co-founding member of the initial version of Univa Grid Engine at Univa with a long history in developing core components of Grid Engine at Sun Microsystems, Oracle, and Univa. Daniel is a strong supporter of open standards. At UberCloud he is committed to “Pave the way to HPC as a Service.”

Terry Smith is Director of Global Partner Solutions at SUSE. Terry is an open source enthusiast and entrepreneur with decades of experience delivering innovations through the power of many. At SUSE, Terry leads the SUSE One Partner Solution Stacks program to deliver co-innovations that arm organizations with the capabilities and agility they need to solve real-world challenges and build a better tomorrow.