Gene Amdahl

principal papers

hardware

software

keywords

see also

related subjects

Achievement

One of the original architects of the business mainframe computer, including IBM’s System/360 computer line(1), Amdahl started the IBM-compatible market when he left IBM to found Amdahl Corporation. Amdahl’s work has been called brilliant and genius by his peers. The Times of London named him one of the "1,000 Makers of the 20^th Century" in 1991, and mainframe magazine Computerworld considered Amdahl one of the 25 people "who changed the world." He is the founder of four companies, Amdahl Corporation(2), Trilogy Systems (now part of Elxsi Corporation), Andor Systems, and Commercial Data Servers (CDS)(3).

Biography

Gene Myron Amdahl was born in South Dakota in 1922. After serving two years in the U.S. Navy during World War II, where he learned electronics, and taking a course in computer programming, he received a bachelors degree in engineering physics at South Dakota State University in 1948. In 1952 he completed his doctorate in theoretical physics at the University of Wisconsin, where he designed his first computer, the Wisconsin Integrally Synchronized Computer (WISC).

He began his career with IBM in 1952, and became the chief design engineer of the IBM 704. In 1955, Amdahl worked with others to design the Datatron, which led to a computer called the Stretch, and eventually became the IBM 7030, a computer that used the new transistor technology. In 1956, after just four short years with IBM, Amdahl became unhappy with the company and quit. After five years of working for other computer companies, he returned to IBM in 1960.

During the 1960s, Amdahl gained recognition as the principle architect of IBM’s impressive System 360 series of mainframe computers. The IBM System 360 was based on the Stretch, which Amdahl had worked on in 1955. The 360 series was one of the greatest success stories in the computer industry and became the main ingredient to IBM’s enormous profitability in the late 1960s.

Leaving IBM...again

Amdahl became an IBM Fellow and was able to pursue his own research projects. In 1969, he was director of IBM’s Advanced Computing Systems Laboratory in Menlo Park, California. He recommended that the laboratory be shut down, which IBM did, and presented his ideas about the internal barriers that prevented IBM from shooting for the high end of computer development. Although his ideas were accepted, IBM executives refused to change policies and Amdahl left IBM again.

In 1970, Amdahl formed his own company, Amdahl Corporation, in Sunnyvale, California. His plan was to compete head-to-head with IBM in the mainframe market. Most industry analysts considered this to be career suicide and gave his start-up company very little chance of surviving. But survive it did, and actually prospered. Instead of creating a rival system to IBM, Amdahl created discounted computers that could be substituted for name brand models and run the same software. Basically, he designed the first computer clones, known then as "plug-to-plug compatibles." Amdahl became the most celebrated entrepreneur in the computer industry for awhile. The only major criticism that was raised about Amdahl Corporation at the time was that Amdahl took start-up money from Fujitsu Ltd. of Japan in exchange for American mainframe technology.

In 1975, Amdahl Corporation shipped its first computer, the Amdahl 470 V/6. Over the next few years, Amdahl and IBM leap-frogged each other with faster, smaller, and cheaper computers. In 1979, Gene Amdahl began moving away from Amdahl Corporation when he resigned his post as chairman. He became chairman emeritus for less than a year, leaving Amdahl Corporation in 1980 to found Trilogy Systems Corporation.

With the success of Amdahl Corporation, Amdahl had no trouble interesting investors in this new company and easily raised $230 million in start-up money. Again, his plan was to compete with IBM, and also Amdahl Corporation, in the high-end mainframe computer market. In addition, Amdahl planned to completely redesign the semiconductor chips that powered the computers. His dream was to combine the functions of 100 separate chips onto one superchip that would work faster and more efficiently than the multiple chips.

Trilogy's Misfortune

Unfortunately, Trilogy was hounded by disasters. Torrential rains delayed construction of the chip plant, then invaded the air conditioning, destroying the clean room atmosphere and all the chips currently being created. At this point, Amdahl had spent one-third of the start-up money with nothing to show for it. To save Trilogy, Amdahl spent the remainder of the money to acquire Elxsi Corporation, a computer manufacturer, in 1985. The new company continued to flounder and never achieved great success. In 1989, Amdahl stepped down as chairman of Elxsi to devote more time to his next venture.

In 1987, Amdahl founded his third company, this one called Andor Systems after the "and" and "or" logic gates of computer circuitry. This time his aim was to build computers that would compete with IBM’s smaller mainframes. Industry analysts uniformly gave the company very little chance of success. But Amdahl felt he had an edge -- he could make small mainframe computers more cheaply than IBM. He could use new technology that allowed him to pack the computer’s central processor onto one board, rather than the several used by IBM, and he redesigned the compiler to work more quickly and efficiently. These innovations allowed Andor’s computers to take up less space and generate less heat, a distinct advantage to customers who no longer would need giant air-conditioned rooms in which to place their computers.

But Andor was plagued by bad chips, causing a delay of almost two years before the first computers hit the market. Meanwhile, IBM came out with its own midsize computer using some of the same technology employed by Andor. To survive, Andor had to come up with other peripheral products that it could quickly get on the market. But Andor never achieved the success it was after with the small mainframes, and in 1991 it had scaled back products to include only a data backup system. By 1994, the company had yet to turn a profit. Eventually, the company declared bankruptcy.

The Main-frame Devotee

But Gene Amdahl was not ready to give up. In 1996, at the age of 74, he started his fourth company, this one called Commercial Data Servers (CDS). Through CDS, Amdahl intends to distribute IBM-compatible, PC-based mainframes that use cryogenically-cooled CMOS processors and a new processor design that he created. CDS is targeting its products at companies that need the capabilities and selling price of a smaller mainframe, a market that CDS believes IBM and other manufacturers aren’t serving adequately.

Gene Amdahl continues his quest to merge mainframe technologies with the more popular PC technology. Though many find these two areas incompatible (mainframe means centralized, controlled computing; PCs are for individual computing), Amdahl won’t give in to those who believe mainframes are dinosaurs that have outlived their usefulness. And, apparently he doesn’t intend to ever give up.

Bibliography

Who’s Who in America

Levine, Jonathan B. "Gene Amdahl tries for two out of three" Business Week, June 27, 1988

Pare, Terence P. "Lions in winter" Fortune, July 4, 1988

"Elxsi names chairman" The New York Times, March 16, 1989, p.D14

Pitta, Julie "Strike two?" Forbes, December 9, 1991

Nash, Jim "Gene Amdahl: mainframe guru still driven, still a believer in vision" The Business Journal, January 10, 1994

Hast, Adele, Ed., International Directory of Company Histories, Volume III, St. James Press, 1991

"Cryogenically-cooled CMOS systems will come out of Amdahl" IBM System User International, July 19, 1996

Courtesy

Chronology

Honors and awards