Professor leading way for Rosetta
From Lawrence Journal-World
By Mark Fagan
His computer language may be taken from a 2,198-year-old slab of black basalt, but Perry Alexander's technical work is anything but ancient.
The associate professor of electrical engineering and computer science at Kansas University is leading the charge to develop Rosetta, a systems design language that one day could help build a new foundation for making faster, stronger and more efficient electronics in everything from wireless phones and personal digital assistants to car engines and climate-control systems.
The language, chiseled from KU research, is envisioned to make microprocessor design more efficient in terms of both time and cost. Designers can create their components virtually by computer, without having to build expensive prototypes or reconfigure production lines once manufacturing has begun.
"When engineers design different things, they use different languages not English and French, but different kinds of mathematics," Alexander said. "For a cell phone, the people who design the antennae the signal processing the power consumption they're all speaking different languages.
"Rosetta provides language support from bringing all those models together. They're still written in their different languages, but Rosetta brings those different languages those different ways of thinking together so that you can look at all the models integrated."
Like its ancient namesake the Rosetta Stone, whose identical inscriptions in three languages unlocked the mysteries of Egyptian hieroglyphics the computer code serves as a universal translator for design engineers, whose work creates the microprocessors that control virtually all electronic products.
The ramifications of Alexander's work on West Campus could be staggering. The market for designing tools that allow for creation of microprocessors and other electronic components is more than $4 billion a year, said Tim Johnson, executive director for KU's Information and Telecommunication Technology Center.
And before Rosetta, he said, nobody has been able to get the world of competing design languages to communicate fluently on the same page.
"Rosetta is laying a foundation, a common framework, so people can design circuits and chips. It provides the underlying bedrock for the industry," Johnson said. "There will be other companies to develop the tools, the services and the training and this is a worldwide industry."
KU and Lawrence could be major benefactors of the Rosetta project, Johnson said. The language recently was donated to Accellera, a California-based nonprofit industry group working on behalf of dozens of designers, manufacturers and software developers, including Intel, Motorola, Texas Instruments, Cadence and Synopsys.
With Alexander leading the group's standards committee for Rosetta, Johnson said, Lawrence will continue to have some of the language's foremost authorities on campus. If the language eventually is adopted as an industry standard, the Lawrence area could become home to small companies aiming to build development, training or service-oriented opportunities out of Rosetta.
"There's not really anyone out there who knows it, because this is a new language," Johnson said. "We think the opportunity is great, because the knowledge base is here. We certainly feel like there's a large market and commercial potential. We are looking at what's the best way to transition the language out so that it will be adopted by the industry."
Johnson said several options were being considered, including licensing opportunities or possible companies that could lead to initial public offerings.
"The industry as a whole is a several billion-dollar industry," he said. "It's just a matter of market share."
For his part, Alexander is busy working on the details. Along with a team of graduate students Garrin Kimmell, Cindy Kong, Brandon Morel, Zhongjun Wang and Satyanarayana Kakarlamudi Alexander is connecting with other academics and professionals around the world, from Australia to Washington, D.C.
He knows their collective project can make a difference. Designers looking to boost battery power in a wireless phone, for example, need to know how design changes would affect other components, which could add to the overall price.
"It helps you make business decisions, to make tradeoffs," Alexander said, noting that Rosetta cuts the number of mistakes and, most notable, helps speed the process.
"The most critical thing to my customers, in electronic design automation, is time time to market. If they get their cool new thing out on the street first, they win."
And Alexander knows his team's work is going into a fertile business field. Only about 0.5 percent of the world's microprocessors are used in desktop computers, leaving the rest to be installed in an ever-expanding lineup of wireless phones, ATM machines, DVD players, modern appliances and more.
"A new Mercedes Benz will have more than 50 microprocessors on board," he said, for such systems as controlling antilock brakes. "There are now more microprocessors on Earth than there are people.
"There are four microprocessors for every person, and we anticipate that going to 10 to 1 in about five years. So what we're doing is a big deal."
For more information, contact ITTC.