Global Sources
EE Times-Asia
Stay in touch with EE Times Asia
EE Times-Asia > EDA/IP

Augmenting Korea's non-memory semiconductor industry

Posted: 29 May 2003 ?? ?Print Version ?Bookmark and Share

Keywords:memory? education? soc? fab? training?

Kyung: "Amaze the world as we did with memory."
Statements like "the strength of the engineering sector translates directly to a nation's competitiveness," and "a competent engineer can make his nation rich" wouldn't come from a mediocre. The mind behind those persuasive accounts is Professor Kyung Jong-Min, a genius well-recognized for his leadership in Korea's R&D efforts in non-memory semiconductor sector. One cannot talk about the development of Korea's semiconductor industry, especially non-memory sector, without mentioning his efforts in the field.

Kyung, who choose to become an engineer, thanks to the strong persuasion of his father, says that engineering is an honest and dignified profession. Being an engineer allows you to bring joy to people by creating something, he said. In 1981, he went to the United States for a two-year stint with Bell Laboratories. He did not spend much of his career working for semiconductor companies, but in education as a professor in the electronic engineering department at the Korean Advanced Institute of Science and Technology (KAIST).

After 20 years in the academe, Kyung has already been recognized both domestically and internationally. He received the KAIST Research Award, one of the design automation institute's highest honor, as well as the Best Treatise Awards from DAC (Design Automation Conference) and ICSPAT (International Conference on Signal Processing Applications and Technology).

These accolades fueled Kyung's desire to develop further the nation's semiconductor industry. In his office room, Kyung proudly showed the construction plan for KAIST's nano fab complex. "That is where we will build the nano-SoC laboratory to train highly-capable chip designers and amaze the world with our non-memory technology, as we did with memory," he said.

Profession with a mission

Being one of the few people who saw through the importance of non-memory semiconductor industry in the mid-1990s, Kyung said that he believes in prompting himself to constantly pursue new things and act with aggressive attitude. "This job requires you to do more than your best; there can be no satisfaction bounds. At first I was eager to write papers as a professor, but while doing 386 projects, I had such a major turn of mindset that I wanted to do things that mean something."

Kyung never seems to run out of ideas to find solutions to Korea's worrisome engineering sector. This time, his focus is on the recent trend of Korean students not taking up engineering. He showed concern for Korea's educational policy that lacks solid foundation in science and technology. "Good engineering students are society's valuable asset and our hope for the future. Policy of granting scholarships without long-term vision for education contributes only to the impoverishment of engineers, and aggravating the current trends of students avoiding the engineering segment," he said.

"In terms of national competitiveness, our future hope lies in SoC design technology considering the demand of IT industry and the fact that the foundation of semiconductor system business is still there. From seeking out new applications to implementing deep-submicron technology, what we need are a few capable EEs rather than an abundant number of mediocre engineers," he said.

To address the social aspects resulting in the avoidance of the engineering segment, "We need to present an overall roadmap showing which levels of work among various areas of scientific technologies we are going to develop and to what extent. This means that we need to focus on high value-added areas such as applications and system design," Kyung emphasized.

"We should be equipped with expertise that allows us to find out new applications consumers want, and efficiently implement these ahead of competing countries. At the core of this is SoC. As its market cycle is short, we need a national SoC laboratory where we can do all the training, research and business in one space. Running this kind of SoC laboratory in the future is my personal ambition," he added.

Moving away from memory

For the last few years, DRAM, a single item in the semiconductor industry, with its incredible record of annual profit, contributed a lot to Korea's export revenue as well as enhancing its technological ability and pride. But it is also a disadvantage that Korea's industry is tilted greatly toward a single item of memory chip. The memory chip business has become an effortless segment which anyone can take on with enough investment in facilities.

Meanwhile, competitive superiority in electronics, automobile and communication industry depends on non-memory technology. And in the non-memory business, having good design talents play a more important role than investment in facilities. Colleges could not provide the comprehensive education for design skills as proper education meant the students themselves needed to design chips. The prices of design tools are too high.

This prompted professor Kyung to organize Integrated Circuit Design Education Center (IDEC) with the assistance of government and private sectors, aiming at nurturing high-class talents for semiconductor design in colleges all over the country. Having run the IDEC for eight years now, Kyung personally considers it as one of his most remarkable projects, along with the development of a Korean microprocessor.

Kyung has been concentrating on KAIST research and educational activities, mainly on the design of non-memory semiconductors including microprocessor and DSP chips. In 1996, he took on a national project together with Hyundai Electronics and succeeded in developing a microprocessor compatible with Intel's 386, 486 and Pentium chips.

At that time, nobody dared to develop microprocessors. But he succeeded in an area where others deemed impossible. "The instruction sets of 386 were so different from those of 286; hence, it was very hard to design. But my goal in the project was like this - if we take out Intel chip from the PC, insert our own chip into it and all the software in the world runs all right, then our project is successful. If not, it has failed. We worked under the object as simple as that. Even though we built the chip based only on the Intel manual, without any other special information, our prototype operated properly by the time our project was finished," he said.

Regrettably, the microprocessor did not make it to commercialization due to troubling issues with licensing and patent. But it was a memorable event worth recording in Korea's history of semiconductor design technology.

Looking ahead

IDEC, having started as an effort from the Ministry of Commerce, Industry, and Energy (MOCIE) to establish the basis of industrial technology, is nearing its second stage. Starting next year, it will be under a different government department. Kyung is worried that it has not yet determined which body would provide support, making its future management somewhat unstable.

"If IDEC stops doing its business, the loss would be too great. To compete with markets such as mainland China, Taiwan and India, we must continue investing in the engineering segment," he emphasized.

He stressed the importance of upgrading engineering skills and nurturing new talents in the country. What the industry needs, he said, are engineers capable of finding and solving new problems.

"When I began working with IDEC, I knew that this was the challenge I cannot afford to fail. From the moment the plans started to nurture semiconductor designers at MOCIE and the forum, to the time IDEC was assigned to KAIST, I asserted my belief that training good engineers will make Korea's future bright. It is my greatest reward to see all the striving students being given opportunity by IDEC to lead affluent and fruitful lives," he concluded.

- Lee Ju-Yeun

Electronic Engineering Times - Asia

Article Comments - Augmenting Korea's non-memory semico...
*? You can enter [0] more charecters.
*Verify code:


Visit Asia Webinars to learn about the latest in technology and get practical design tips.

Back to Top