High Performance (Potato) Chips

By Michael Feldman

May 5, 2006

“I’m going to be talking about things that are very familiar to people,” said Tom Lange, Director of Modeling and Simulation at Procter & Gamble (P&G).

Not the kind of introduction you normally think of when someone speaks about high performance computing applications. But this is exactly what Tom Lange talked about at the High Performance Computing and Communications (HPCC) Conference in Newport, Rhode Island a few weeks ago. His presentation was titled “The Aerodynamics of Pringles.”

Tom Lange has spent over 27 years at Procter & Gamble, modeling products, processes and production systems — everything from how the aerodynamics of potato chips optimizes production to how baby size affects diaper leakage. Although P&G has really only used high performance computing for the last 10 years or so, its origins go back to the late 70s.

“When I joined Procter & Gamble in 1978, we had high-end IBM 360/370 kinds of computers that we used to solve statistics problems,” said Lange. “Our first finite element analysis kind of problem — something that would look more familiar to a supercomputing person today — we solved using a Boeing computer in the middle of the 1980s. So our exploration of the use of simulations to improve our ability to innovate for the consumer is a legacy that is not just a few years old, but in fact more like 15 years old.”

Today, P&G has a fairly typical setup for commercial users of high performance computing. Lange said they have a heterogeneous computing environment — a shared memory SGI Altix system and a multi-hundred-node cluster. Choosing which system to use depends on their suitability for the specific type of modeling/simulation application.

As far as software goes, P&G gets its codes from a variety of sources. They use software packages from ISVs like Abacus, Fluent and LS-Dyna. Most of P&G’s proprietary code is implemented with user-defined functions within these packages. Lange calls this his “commercial-plus” strategy. At P&G, they have not attempted to maintain internal codes.

P&G also uses some national laboratory codes from both LANL and Sandia National Labs. “The same weapons code used at Los Alamos for more sophisticated purposes is used for combustion code in automotive applications and at P&G for paper products manufacturing,” said Lange.

Procter & Gamble tells its story

Unlike its competitors, P&G’s been publicizing how it uses high performance computing technology for a few years now. Other companies have been much more reticent to share their HPC story with the masses. Even Lange admits this story would not have told at P&G in the 1980s. But the nature of product manufacturing has changed.

“We’re in a global competition for ideas,” said Lange. “There’s no illusion at Procter and Gamble that it’s the only place where smart things happen. Since that illusion is not there, our willingness to say what we do know gives us the hope we’ll learn from others. If we’re just sitting in the back hiding, not saying anything, that doesn’t improve our innovation.”

Procter & Gamble does appear to have a more strategic focus on using HPC technology than its competitors. Lange’s position — the director of modeling and simulation — may be hard to find at other companies that produce package goods. Although modeling may have been used to help with product and package design at P&G ten or fifteen years ago, it wasn’t seen as a critical asset. But today, Lange believes there is an increasing awareness to use this technology to develop and improve products. This mirrors what has happened in other sectors — defense, electronics, automotive, aerospace, oil & gas — in the last decade or so.

Lange believes his willingness to speak at conferences like HPCC helps him connect with others in government and industry that deal with similar types of problems. He is hoping to develop some good relationships at the conference, leading to possible future collaborations. Lange uses events such as these to get to know his counterparts in other organizations.

“I know my counterparts at Chrysler, I know my counterparts at Dreamworks, I know my counterparts at Morgan-Stanley,” said Lange. “I would have never met those individuals if I hadn’t been involved in things like [HPCC]. In a lot of ways they all have similar jobs to mine. They’re trying to bring computing to their innovation process.”

Lange believes that collaboration between the defense, automotive industry, and package goods industry is quite possible. For example, P&G models many of the properties of skin to develop the interaction of its lotion products. Those models could be relevant for a crash test simulation at Ford Motor Company or a battlefield armor protection simulation for the Army.

“In my world I’m worried about wrinkles and freckles,” said Lange. “I’m just trying to make everyone’s life just a little better. But the science and engineering of making everyone’s life a little better has an amazing similarity to what are some of the more complex problems in safety and defense.”

High Performance Pringles

In general, Procter and Gamble use high performance computing modeling to design consumer package goods for a variety of its products: Ivory, Pringles, Charmin, Downy, Tide, Crest, Mr. Clean, Pampers, and a whole range of Hugo Boss products. A fairly recent success story is the Folgers Coffee plastic canister, which features the so-called “Aroma Seal.”

“There’s a lot of complex science and engineering associated with that particular container,” said Lange.

He explains that structural integrity is especially important for hermetically sealed packages. This type of container must be able to withstand pressure changes in elevation when they’re being transported — for example, during shipping, when the product is being driven over 11,000-foot mountain passes. Metal containers are very resistant pressure changes. But metal has drawbacks in maintaining the flavor profiles of foods, such as coffee, whose aroma is a result of its volatile oils. Metal does not react well with those volatile oils, so the coffee flavor tends to degrade over time.

Plastic, on the other hand, is better at preserving the coffee flavor profile. However plastic is not as good at maintaining its structural integrity when undergoing pressure changes during transport. Lange said this can be overcome if you just make the plastic really thick, but this is not very practical from a consumer acceptance and environmental point of view. So the challenge was to design a plastic container that would be both strong and practical for the consumer. For this, Procter and Gamble had to resort to sophisticated computer-aided engineering.

“That plastic coffee canister — the Aroma Seal package — would not exist without modeling,” said Lange. “Packaging, in general, is where this [modeling] gets applied — whether you’re talking about a Tide bottle or any of our liquid products.”

At P&G, product modeling is used to design a range of properties associated with a package, including its manufacturability, its strength and it resistance to leakage. In some cases, modeling is used to create more efficient packaging, so that fewer raw materials are used. This benefits both the manufacturer, because it is less expensive to produce, and the consumer, because its lighter, more compact and friendlier to the environment.

According to Lange, their paper products, including disposable diapers, toilet paper and paper towels is another area where a lot of modeling takes place. Also, substrate-based products such as Swiffer, Bounce, Thermocare have also benefited from high performance computing, employing chemoinformatics and molecular mesoscale modeling to predict the behavior of liquid solutions. Lange said that none of these products would be on the store shelves without modeling.

And then there’s Pringles. One of the reasons the aerodynamics of Pringles is so important is because the chips are being produced so quickly that they are practically flying down the production line.

“We make them very, very, very fast,” said Lange. “We make them fast enough so that in their transport, the aerodynamics are relevant. If we make them too fast, they fly where we don’t want them to, which is normally into a big pile somewhere. And that’s bad.”

Lange notes that the aerodynamics of chips is also important for food processing reasons. In this case, the aerodynamic properties combine with the food engineering issues, such as fluid flow interactions with the steam and oil as the chips are being cooked and seasoned.

Future Applications

Lange thinks that he will be able to use more advanced codes, such as human biomechanical modeling, on next-generation computers. At P&G, he would like to apply biomechanical modeling to design more user-friendly packaging. To the degree Procter and Gamble’s products interface better with the full range of humanity, the more likely he’s going to able to deliver a preferred product in the marketplace.

Lange describes one possible application of this from his own experience. He said he noticed that his mother-in-law, who has arthritis, leaves tops ajar or the caps off on a variety of containers around her home, because it’s too painful for her to continually open and close them.

“It’s a classic engineering dilemma, said Lange. “How do I make something that never leaks but opens easily? Introducing the human into this, in a full biomechanical way, is a complicated problem. It puts a huge demand on computing.”

Lange said that if they had more computing power, they could also perform much finer-grained molecular modeling. For example, they could simulate the nanoscale behavior of liquids. With this capability they would be able to predict the stability and opacity properties of different liquid solutions. Today he can only address those problems with very simple mesoscale representations.

Lange thinks it’s a shame when he occasionally hears his counterparts in the aerospace and automotive sectors say their systems are fast enough today — that no more computing power is really needed. He believes there are problems in all engineering domains that have yet to be addressed because of a lack of computing capability.

“My appetite for computing is insatiable,” admitted Lange. “For every factor of ten that Moore’s Law gives me, I can make use of every bit of it!”

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Data Vortex Users Contemplate the Future of Supercomputing

October 19, 2017

Last month (Sept. 11-12), HPC networking company Data Vortex held its inaugural users group at Pacific Northwest National Laboratory (PNNL) bringing together about 30 participants from industry, government and academia t Read more…

By Tiffany Trader

AI Self-Training Goes Forward at Google DeepMind

October 19, 2017

DeepMind, Google’s AI research organization, announced today in a blog that AlphaGo Zero, the latest evolution of AlphaGo (the first computer program to defeat a Go world champion) trained itself within three days to play Go at a superhuman level (i.e., better than any human) – and to beat the old version of AlphaGo – without leveraging human expertise, data or training. Read more…

By Doug Black

Researchers Scale COSMO Climate Code to 4888 GPUs on Piz Daint

October 17, 2017

Effective global climate simulation, sorely needed to anticipate and cope with global warming, has long been computationally challenging. Two of the major obstacles are the needed resolution and prolonged time to compute Read more…

By John Russell

HPE Extreme Performance Solutions

Transforming Genomic Analytics with HPC-Accelerated Insights

Advancements in the field of genomics are revolutionizing our understanding of human biology, rapidly accelerating the discovery and treatment of genetic diseases, and dramatically improving human health. Read more…

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Cluster Competition coverage has come to its natural home: H Read more…

By Dan Olds

Data Vortex Users Contemplate the Future of Supercomputing

October 19, 2017

Last month (Sept. 11-12), HPC networking company Data Vortex held its inaugural users group at Pacific Northwest National Laboratory (PNNL) bringing together ab Read more…

By Tiffany Trader

AI Self-Training Goes Forward at Google DeepMind

October 19, 2017

DeepMind, Google’s AI research organization, announced today in a blog that AlphaGo Zero, the latest evolution of AlphaGo (the first computer program to defeat a Go world champion) trained itself within three days to play Go at a superhuman level (i.e., better than any human) – and to beat the old version of AlphaGo – without leveraging human expertise, data or training. Read more…

By Doug Black

Student Cluster Competition Coverage New Home

October 16, 2017

Hello computer sports fans! This is the first of many (many!) articles covering the world-wide phenomenon of Student Cluster Competitions. Finally, the Student Read more…

By Dan Olds

Intel Delivers 17-Qubit Quantum Chip to European Research Partner

October 10, 2017

On Tuesday, Intel delivered a 17-qubit superconducting test chip to research partner QuTech, the quantum research institute of Delft University of Technology (TU Delft) in the Netherlands. The announcement marks a major milestone in the 10-year, $50-million collaborative relationship with TU Delft and TNO, the Dutch Organization for Applied Research, to accelerate advancements in quantum computing. Read more…

By Tiffany Trader

Fujitsu Tapped to Build 37-Petaflops ABCI System for AIST

October 10, 2017

Fujitsu announced today it will build the long-planned AI Bridging Cloud Infrastructure (ABCI) which is set to become the fastest supercomputer system in Japan Read more…

By John Russell

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Intel Debuts Programmable Acceleration Card

October 5, 2017

With a view toward supporting complex, data-intensive applications, such as AI inference, video streaming analytics, database acceleration and genomics, Intel i Read more…

By Doug Black

Reinders: “AVX-512 May Be a Hidden Gem” in Intel Xeon Scalable Processors

June 29, 2017

Imagine if we could use vector processing on something other than just floating point problems.  Today, GPUs and CPUs work tirelessly to accelerate algorithms Read more…

By James Reinders

NERSC Scales Scientific Deep Learning to 15 Petaflops

August 28, 2017

A collaborative effort between Intel, NERSC and Stanford has delivered the first 15-petaflops deep learning software running on HPC platforms and is, according Read more…

By Rob Farber

How ‘Knights Mill’ Gets Its Deep Learning Flops

June 22, 2017

Intel, the subject of much speculation regarding the delayed, rewritten or potentially canceled “Aurora” contract (the Argonne Lab part of the CORAL “ Read more…

By Tiffany Trader

Oracle Layoffs Reportedly Hit SPARC and Solaris Hard

September 7, 2017

Oracle’s latest layoffs have many wondering if this is the end of the line for the SPARC processor and Solaris OS development. As reported by multiple sources Read more…

By John Russell

US Coalesces Plans for First Exascale Supercomputer: Aurora in 2021

September 27, 2017

At the Advanced Scientific Computing Advisory Committee (ASCAC) meeting, in Arlington, Va., yesterday (Sept. 26), it was revealed that the "Aurora" supercompute Read more…

By Tiffany Trader

Google Releases Deeplearn.js to Further Democratize Machine Learning

August 17, 2017

Spreading the use of machine learning tools is one of the goals of Google’s PAIR (People + AI Research) initiative, which was introduced in early July. Last w Read more…

By John Russell

GlobalFoundries Puts Wind in AMD’s Sails with 12nm FinFET

September 24, 2017

From its annual tech conference last week (Sept. 20), where GlobalFoundries welcomed more than 600 semiconductor professionals (reaching the Santa Clara venue Read more…

By Tiffany Trader

Graphcore Readies Launch of 16nm Colossus-IPU Chip

July 20, 2017

A second $30 million funding round for U.K. AI chip developer Graphcore sets up the company to go to market with its “intelligent processing unit” (IPU) in Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Responds to Google TPU Benchmarking

April 10, 2017

Nvidia highlights strengths of its newest GPU silicon in response to Google's report on the performance and energy advantages of its custom tensor processor. Read more…

By Tiffany Trader

Amazon Debuts New AMD-based GPU Instances for Graphics Acceleration

September 12, 2017

Last week Amazon Web Services (AWS) streaming service, AppStream 2.0, introduced a new GPU instance called Graphics Design intended to accelerate graphics. The Read more…

By John Russell

EU Funds 20 Million Euro ARM+FPGA Exascale Project

September 7, 2017

At the Barcelona Supercomputer Centre on Wednesday (Sept. 6), 16 partners gathered to launch the EuroEXA project, which invests €20 million over three-and-a-half years into exascale-focused research and development. Led by the Horizon 2020 program, EuroEXA picks up the banner of a triad of partner projects — ExaNeSt, EcoScale and ExaNoDe — building on their work... Read more…

By Tiffany Trader

Delays, Smoke, Records & Markets – A Candid Conversation with Cray CEO Peter Ungaro

October 5, 2017

Earlier this month, Tom Tabor, publisher of HPCwire and I had a very personal conversation with Cray CEO Peter Ungaro. Cray has been on something of a Cinderell Read more…

By Tiffany Trader & Tom Tabor

Cray Moves to Acquire the Seagate ClusterStor Line

July 28, 2017

This week Cray announced that it is picking up Seagate's ClusterStor HPC storage array business for an undisclosed sum. "In short we're effectively transitioning the bulk of the ClusterStor product line to Cray," said CEO Peter Ungaro. Read more…

By Tiffany Trader

Intel Launches Software Tools to Ease FPGA Programming

September 5, 2017

Field Programmable Gate Arrays (FPGAs) have a reputation for being difficult to program, requiring expertise in specialty languages, like Verilog or VHDL. Easin Read more…

By Tiffany Trader

IBM Advances Web-based Quantum Programming

September 5, 2017

IBM Research is pairing its Jupyter-based Data Science Experience notebook environment with its cloud-based quantum computer, IBM Q, in hopes of encouraging a new class of entrepreneurial user to solve intractable problems that even exceed the capabilities of the best AI systems. Read more…

By Alex Woodie

HPC Chips – A Veritable Smorgasbord?

October 10, 2017

For the first time since AMD's ill-fated launch of Bulldozer the answer to the question, 'Which CPU will be in my next HPC system?' doesn't have to be 'Whichever variety of Intel Xeon E5 they are selling when we procure'. Read more…

By Dairsie Latimer

  • arrow
  • Click Here for More Headlines
  • arrow
Share This