Big-name manufacturers such as GM, Boeing, DreamWorks and Eli Lilly are using high-performance computing to design better products, sure, but they’re also saving money by implementing the computers’ ability to simulate real-world testing. Take the automaker GM, for example: in order to test the safety of its cars, it needs to crash them. With some cars costing more than $300,000 (and crash test dummies costing up to $100,000), it makes sense to keep the number of crash tests to a minimum. That’s where supercomputers come in. Despite the hefty costs associated with purchasing and maintaining high-end supercomputers, it still makes financial sense.
The cost-savings HPC affords by reining in the need for physical testing is the subject of an article at Bloomberg Business Week by Rachel King. HPC’s reach is extensive, running the gamut from vehicle design to animation to drug development to seismic imaging, as King illustrates:
Engineers at GM use high-performance computers to simulate the new 2011 Chevrolet Cruze, while Boeing (BA) used them in developing the 787 Dreamliner. These machines help animators at DreamWorks Animation SKG (DWA) render movies such as Shrek and Kung Fu Panda, while Eli Lilly & Co. (LLY) scientists use them to research new pharmaceuticals. Chevron (CVX) used high-performance computing to do seismic imaging that led to the discovery of new reservoirs of oil in the Gulf of Mexico and Speedo International took advantage of it to model the swimsuit Michael Phelps wore at the 2008 Olympics.
Experts cited in the article call the technology “game-changing,” and say that it ”increases safety and overall vehicle performance.”
Every technology has its drawbacks, however, and supercomputing is not an exception. The complexity of the software can create barriers to use and there are cost and environmental concerns related to keeping all those processors cooled.
Virtual testing can never completely obviate the need for physical testing, but together the two methods work synergistically to advance the design and manufacturing processes, with the virtual testing pointing the way and the physical testing providing the verification.