Download PDF
Altair > Case Studies > Ford Motor Company: Building an Efficient HPC Infrastructure
Altair Logo

Ford Motor Company: Building an Efficient HPC Infrastructure

Technology Category
  • Infrastructure as a Service (IaaS) - Cloud Computing
  • Sensors - Temperature Sensors
Applicable Industries
  • Automotive
  • Cement
Applicable Functions
  • Product Research & Development
  • Quality Assurance
Use Cases
  • Construction Management
  • Infrastructure Inspection
Services
  • Hardware Design & Engineering Services
  • Testing & Certification
The Challenge
Ford's Numerically Intensive Computing Department (NIC) had built a substantial heterogeneous High-Performance Computing (HPC) environment over the years, combining both capacity and capability. This environment included Beowulf clusters based on Xeon, Itanium, and Alpha processors, SGI Origin and Altix servers, IBMP650 capacity, and large SMP Cray systems. While this infrastructure enabled NIC to process compute-intensive jobs in a timely manner, it also resulted in a complex infrastructure of platforms and applications. Additionally, NIC faced complexity on the solver side, running many application versions, none of which ran on all architectures. The challenge was to find a solution that could efficiently manage this complex infrastructure and provide a simple tool for users.
About The Customer
Ford Motor Company is a leading automotive manufacturer, producing cars, trucks, vans, and SUVs that are safer, quieter, and more comfortable than ever. Many of the intangibles that contribute to Ford's quality flow from the innovative use of high-performance computing (HPC) techniques. At Ford’s Numerically Intensive Computing Department (NIC) in Dearborn, Michigan, engineers run simulations in codes such as NASTRAN and LS-DYNA for predictive analysis of cylinder cooling, wind noise, vibration, ride quality, crashworthiness, durability, and other characteristics that contribute to industry-leading automotive design.
The Solution
NIC adopted PBS Professional to manage its complex hardware and software with job attributes. The previous workload manager had reached its scalability limits and could not efficiently handle NIC's increasing processor count. After evaluating commercially available workload management software, NIC chose PBS Professional due to its cost, flexibility, and availability of source code. The availability of PBS Professional source code allowed NIC to run the PBS Professional server infrastructure and develop a custom scheduling package that runs on top of it. PBS Professional infrastructure gave NIC the freedom to move in a technology market where processors are rising in performance and dropping in cost. It allowed NIC to bring in the latest technologies to ensure that Ford is getting maximum value for its hardware dollar, and PBS Professional enabled NIC to plug them in and present them to users as part of a single resource.
Operational Impact
  • The adoption of PBS Professional has allowed NIC to manage its complex infrastructure more efficiently. It has provided the freedom to move in a technology market where processors are rising in performance and dropping in cost. This has enabled NIC to bring in the latest technologies, ensuring that Ford is getting maximum value for its hardware dollar. Furthermore, PBS Professional has enabled NIC to plug these technologies in and present them to users as part of a single resource. This has brought NIC closer to a virtualized approach, allowing engineers to focus more on engineering tasks.
Quantitative Benefit
  • PBS Professional provides freedom to move in a technology market where processors are rising in performance and dropping in cost
  • PBS Professional enables NIC to bring in the latest technologies to ensure that Ford is getting maximum value for its hardware dollar
  • PBS Professional allows NIC to plug in the latest technologies and present them to users as part of a single resource

Related Case Studies.

Contact us

Let's talk!

* Required
* Required
* Required
* Invalid email address
By submitting this form, you agree that IoT ONE may contact you with insights and marketing messaging.
No thanks, I don't want to receive any marketing emails from IoT ONE.
Submit

Thank you for your message!
We will contact you soon.