Just curious: An interview with John Cocke

Abstract

This computer architecture leader's curiosity led him to discover several of the field's most significant advances. lcgcnd in the computer arcliitccture coinmuuity, John Cockc has been involved in tlie design of several machines that have made a tremendous impact ou current processor design, including the IBM Stretch; the Advanced Computer System (ACS); and the 801, RS/6000, and I'owerPC proccssors. Pcrhaps best known as a pionccr of ideas that lcd to reduccd iiistruction set computing (RISC), Cocke is also much admired for a broad intcrest in and understanding of techuology that spans mathematics, compilers, architecture, circuits, packaging, and design automation, to namc a few. In conjunction with his winning the inaugural Seymour Cray Award, Computer visited Cocke in his Westchcstcr, New York, home, located near the T.J. Watson Research Center, where he worked for iiearly four decades until his retirement. Beginnings Computer: Let's start at the bcginning. How and whcn did you initially become acquainted with computers? And when did thcy begin to fascinate you? Cncke: Sullivan Camphell, who worked with IBM for several years, came to Duke after working on Oracle at Oak Ridge, Tennessee. Oraclc was essentially a 40-hit-wide parallcl von Neumann computer like the one at the Institutc for Advanccd Study. I Iiad just graduated and had planncd to spend the summer at Duke working for J.J. Gurgen, 3 mathcmatician who was hired to study which computers the Army might use. I rented a room from Sully Campbell-he was working for Gurgen too-and we drank a lot of becr and spent a lot of time talking about computers. I had studied mathematics but didn't know auythiug about computers until I sat around and talked to Campbell ahout Oracle, and lxiilding faster adders-siinple-minded things like that. The field just interested me. Computer: llow did you come to work for IBM? Cockc: I was thinking of going to work for Arthur U. Little, so I went by thcrc and by GE. I had a friend at IBM-a logician named Brad Dunhan-who took me around to see Steve Dunwell, the hcad of the Stretch project, which was just starting. Dunwell convinccd me that working on Stretch would be interesting-one of the main goals was to make it run fast-and so I joined IBM in 1956. Wc had a very interesting group that included Jim Pomerene-who huilt the CRT memory for tlie Institute machine, thc Jirhnuiac-and Fred Brooks. I had a desk in between them. I learned about Fortran from Irv Ziller, who worked on the original compiler and invented Fortran 3. Our tCam also had Gerry Blaauw and John Fairclough. John later worked as manager of the IBM Hurslcy lab and tlicn for the prime minister, and he was later knighted. I was delighted that the people I met knew something ahout computers hecause I didn't. Subsequently,

IBM's single-processor supercomputer efforts

“…6 9 Cocke said this goal was a reaction to Amdahl, who after returning to IBM had postulated an upper limit of one instruction decode per cycle. 8 In 1965, responding to the announcement of the CDC 6800 (later to become the 7600) and to Seymour Cray's success with a small isolated design team for the 6600, Watson expanded Project Y and relocated it to California near the company's San Jose disk facility. It was now called ACS (Advanced Computer Systems).…”