Visit additional Tabor Communication Publications
March 21, 2008
Software is one of Tom Lange's favorite subjects -- or least favorite, depending on his mood. Lange heads the modeling and simulation group at Procter & Gamble (P&G) and is responsible for enlisting computer technology to help develop the company's vast array of consumer products. He is well-know in the HPC community as an outspoken evangelist for high performance computing in industry.
Last week, Lange spoke at the HPC Horizons Summit in Palm Springs, where a number of industry luminaries provided their perspectives on how users are pushing the envelope of HPC usage. This two-day event was organized by Tabor Communications, the parent company of HPCwire, and provided an opportunity for the HPC digerati to talk about emerging applications and bask in the 85-degree desert sunshine. Lange's presentation was a mostly feel-good story of how one company has used HPC to help create some common everyday products.
With over $76 billion in revenue last year, Procter & Gamble is the largest consumer goods company in the world. The company uses high-end computing to design, test and manufacture a wide variety of consumer products (as well as the packaging they come in). A short list of HPC-enabled products includes Pampers diapers, Downy bottles, Braun shavers, and Pringles potato chips. As in many HPC applications, the idea is to replace physical research and development with computer simulations. When you're talking about designing leak-proof diapers, the advantages of avoiding wetlab conditions become more obvious.
Unlike HPC-crafted products such as commercial airplanes or Formula One race cars, P&G consumer products are produced by the billions. So materials and manufacturing costs are as critical to product design as usability. With an average P&G product price of under $10, there's a huge incentive to minimize packaging and simplify assembly. Usability is still a big challenge since fussy consumers are going to demand an array of conflicting characteristics: Materials must be strong, but soft, even when wet; they must stretch but not break. Liquid mixtures must be easy to dispense, but be thick enough to stay in place when they're applied. Packages must be strong, lightweight, leak-proof, safe to handle, but easy to open.
Something as apparently simple as Tide laundry detergent could require as much computer modeling sophistication as a Boeing 747. For example, liquid detergent may require three distinct modeling applications -- one for soap characteristics, one for the bottle design, and one for filling the bottle on the assembly line. So this single product may suck up different computing resources and require a complex set of software that involves CFD, FEA and CAE codes.
On the positive side, computing is getting cheaper every year. P&G has seen the price of hardware drop from around a $1.50/CPU-hour in 2001 to $0.15/CPU-hour in 2007. Whereas other manufacturers have used the lower-cost FLOPs to reduce IT expenditures, P&G wants to take advantage of the increased price-performance to expand research. With more than 40 individual brands that net over a $500 million each in revenue, there is plenty of incentive to improve the manufacturability and usability of their product set.
According to Lange, the next set of challenges for designing consumer goods will be to inject more realism into the simulations, for example, using nanoscale chemistry modeling to predict the biochemical behavior of skin lotions, or using biomechanical simulations to measure the ability of a child to open a lid. This type of application is within reach today, but usually only on top tier supercomputers. Since industry tends to lag the top systems by a generation or two, the current teraflop systems used by large commercial users like P&G are five or ten years behind the curve, performance-wise.
Lange is actually much less worried about getting enough computing muscle than he is about the software. By 2010, he expects the cost of computing hardware to drop to just a few cents/CPU-hour. But since that hardware will be implemented with lots of parallelism, the current core-based licensing models will put software costs onto a Moore's Law trajectory. Like many users, Lange is frustrated that the advantages of more powerful hardware are being overwhelmed by the increasing cost of the software.
He's not alone. There was plenty of angst expressed about software costs during the HPC Horizons Summit. Both vendors and users see licensing costs as a big impediment to expanding HPC usage. Part of this problem is cultural. People aren't yet used to the idea that software is a much more valuable commodity than hardware, since, up until recently this wasn't the case. Also, in an era when open source is making software tools and operating systems widely available, people can convince themselves that "free" software has no cost associated with it.
Lange admits he doesn't know how the marketplace will resolve this. P&G uses software from both commercial sources, like ANSYS, and DOE national labs, like Lawrence Livermore and Sandia. Up until now at least, Lange's simulation and modeling group has not attempted to maintain codes internally because of manpower costs. Developing in-house codes represents the ultimate in control, but for P&G that would represent a significant shift in computing strategy. Lange admits that the establishment of a large in-house software engineering group is not a good fit with the business culture at P&G, which stresses long-term career paths and promoting from within. The company prefers to concentrate on what it does best -- understanding consumer needs and making the products that fill those needs. Ultimately, P&G would like to be a user of HPC and not a developer. I'm guessing, that attitude reflects the feelings of most commercial users.
At the end of Lange's presentation, he asked a number of tough questions about the business case for software in an increasingly parallel world:
I didn't notice any commercial application software vendors at the HPC Horizons Summit, but it would be great to see some of the ISV leadership weigh in on this discussion. If a company like Procter & Gamble with deep pockets and a obvious commitment to HPC starts to balk at software affordability, it's probably time for the whole community to get serious about this topic.
As always, comments about HPCwire are welcomed and encouraged. Write to me, Michael Feldman, at email@example.com.
Posted by Michael Feldman - March 20, 2008 @ 9:00 PM, Pacific Daylight Time
Michael Feldman is the editor of HPCwire.
No Recent Blog Comments
In quieter times, sounding the bell of funding big science with big systems tends to resonate further than when ears are already burning with sour economic and national security news. For exascale's future, however, the time could be ripe to instill some sense of urgency....
In a recent solicitation, the NSF laid out needs for furthering its scientific and engineering infrastructure with new tools to go beyond top performance, Having already delivered systems like Stampede and Blue Waters, they're turning an eye to solving data-intensive challenges. We spoke with the agency's Irene Qualters and Barry Schneider about..
Large-scale, worldwide scientific initiatives rely on some cloud-based system to both coordinate efforts and manage computational efforts at peak times that cannot be contained within the combined in-house HPC resources. Last week at Google I/O, Brookhaven National Lab’s Sergey Panitkin discussed the role of the Google Compute Engine in providing computational support to ATLAS, a detector of high-energy particles at the Large Hadron Collider (LHC).
May 23, 2013 |
The study of climate change is one of those scientific problems where it is almost essential to model the entire Earth to attain accurate results and make worthwhile predictions. In an attempt to make climate science more accessible to smaller research facilities, NASA introduced what they call ‘Climate in a Box,’ a system they note acts as a desktop supercomputer.
May 22, 2013 |
At some point in the not-too-distant future, building powerful, miniature computing systems will be considered a hobby for high schoolers, just as robotics or even Lego-building are today. That could be made possible through recent advancements made with the Raspberry Pi computers.
May 16, 2013 |
When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
May 15, 2013 |
Supercomputers at the Department of Energy’s National Energy Research Scientific Computing Center (NERSC) have worked on important computational problems such as collapse of the atomic state, the optimization of chemical catalysts, and now modeling popping bubbles.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.