Cloud computing keeps on trucking
![]() |
| Preston Feight, chief engineer, uses cloud computing to redesign Kenworth trucks without making huge investments in technology. |
Most people don't spend much time thinking about mudflaps – those strips of rubber behind a big rig's wheels that repel grime and maybe show a gun-toting Yosemite Sam, warning "Back off!" But by using sophisticated design technology, engineers at truckmaker Kenworth discovered that the little flaps were also a major source of drag. Simply trimming and tapering the flaps will cut about $400 from a typical truck's annual gas bill.
How exactly did Kenworth solve its mudflap quandary? The company, a unit of $15-billion-a-year truck conglomerate Paccar, took advantage of cloud computing. In Kenworth's case, engineers rented time on a supercomputer thousands of miles away; that system helped employees root out assorted gas-guzzling design flaws they might have missed had they relied just on computers at Kenworth's facilities outside Seattle.
Indeed, the engineers used rented computing gear to produce the T660, an aerodynamic truck released in 2007 that ditched the typical bulldog look for a more dolphin-like snout and can get roughly seven miles per gallon. (Don't laugh. That's actually impressive for a semi.)
(INTC) (IBM) (DELL) (HPQ) (AMD) (NVDA) (CRM) (JAVA)
Great article!
I think we'll see much more of this sort of "private cloud" in the future — federation of internal and external IT infrastructure under the control of IT.
Nice the way you tied it to real-world business benefits — something we should all strive to do more of!
I've written several posts on this topic, as I think it's going to be a popular theme in IT thinking going forward.






This is an excellent case study for ways cloud computing might be used. They don't indicate what "super-computer" was used but, one could achieve similar results if they know how to use Amazon EC2, for example, as batch computing infrastructure to solve problems. In fact, this is already happening in other use cases. Use what you need, when you need it, pay as you go, and elastically expand or contract your server footprint when you need.
Kent Langley, CTO
nScaled, Inc.
http://www.nscaled.com