Automotive supplier doubles in size after adopting HPC

National effort aims to make high performance computing available to manufacturing firms

Many small to mid-size manufacturing companies do not use high performance computing (HPC) to create and test potential parts and products virtually because of cost concerns. But one firm that did make the investment in HPC developed a new product line -- and subsequently doubled in size.

The company, L&L Products, in Romeo, Mich., is an automotive supplier that makes proprietary chemicals that can be used in any number of applications. Its products include high-strength adhesives that will bond to any material and structural composites used to strengthen vehicles.

L&L began using high performance computing about six years ago to build a new structural composite line for automotive makers. To accomplish this, it needed to design and test its products in vehicle crash simulations to see how they could be best applied in automobiles. The composite line became a new product line for the company, one that would have been impossible to start without HPC resources, said Steven Reagan, the computational modeling manager at L&L.

Reagan uses HPC by first creating a virtual model of a part to add to a vehicle design. He will run automotive crash simulations on his Linux cluster to test its performance.

"It doubled our business," said Reagan, adding that the firm now makes about $220 million annually and has about 900 global employees.

Before the introduction of HPC, "we didn't compete in this market," said Reagan. "There is no way to compete in this market without that tool."

Reagan is now helping Jon Riley, a former L&L employee who helped to launch the HPC operation, to create a national HPC effort. The goal is to provide technical talent, compute resources and HPC applications to small-and-mid-sized firms.

"If we want to get them (small and mid-sized manufacturers) to be contributors to our national economy, you have to give them the tools to do that," said Reagan. "You have to give them the hardware tools, the software tools, and the people with experience who know how to use them."

Riley, as executive director of design and engineering programs at the National Center for Manufacturing Sciences (NCMS) in Ann Arbor, Mich., is working on a program to create about a dozen centers throughout the U.S. to connect manufacturing firms with HPC resources.

The plan is to establish these centers near universities and national labs to tap into local expertise. NCMS will run the centers and take on collaborating employees -- people who don't work for NCMS, but bring various skills to help manufacturers.

Each center will cost about $15 million to $20 million to establish and will offer portals and a cloud-based set of tools that connect manufacturers with modeling and simulation resources, said Riley. The non-profit NCMS will charge fees for services, with the goal of making the centers self-sustaining after a few years of operation.

The compute resources will be "behind a curtain," and invisible to customers. Potential clients "don't know what InfiniBand is and they don't want to know," said Riley. "What they want to do is get the job done."

Rile, who hopes to have two centers established next year with the help of state and federal funding, believes there are some 300,000 companies that may benefit from such a service.

Earl Joseph, an HPC analyst at IDC, called the NCMS effort a "fantastic idea," saying it is "absolutely right thing to do -- I would just like to see it dramatically larger."

Joseph said the number of manufacturing firms likely to see the most benefit from HPC is probably closer to 25,000. He would like to see a program that can reach at least 2,000 of these companies a year.

Patrick Thibodeau covers cloud computing and enterprise applications, outsourcing, government IT policies, data centers and IT workforce issues for Computerworld. Follow Patrick on Twitter at @DCgov or subscribe to Patrick's RSS feed . His e-mail address is pthibodeau@computerworld.com.

Join the discussion
Be the first to comment on this article. Our Commenting Policies