His most famous accomplishments were the construction of the first operating stored program computer, EDSAC in June, 1949 and the invention of microprogramming, (published in IEEE Spectrum in 1955).
His later innovations never gained commercial importance because computer companies in Europe were not as commercially important as those in the U.S.
He was educated at St., John's College, Cambridge from 1931 to 1934. In 1936 he obtained his Ph.D. from Cambridge for studies in ionospheric propagation of long radio waves. He worked as a lecturer at Cambridge until he was appointed to the head of the new Cambridge computing laboratory in 1945, a post he held until 1980, when he moved to engineering at Digital Equipment Corporation.
The Cambridge laboratory initially had many different computing devices, including a differential analyzer. He obtained a copy of John Von Neumann's prepress description of the EDVAC, a successor to the ENIAC under construction by Presper Eckert and John Mauchly at the Moore School of Electronics. He had to read it overnight because he had to return it and no photocopy facilities existed. He decided immediately that this was the way computing must proceed.
Since his laboratory had its own funding, he was immediately able to start work on a small practical machine, the EDSAC. He decided that his mandate was not to invent a better computer, but simply to make one available to the university. Therefore his approach was relentlessly practical. He used only proven methods for constructing each part of the computer. The resulting computer was slower and smaller than other planned contemporary computers. However, his laboratory's computer, the EDSAC, was finished before any other stored program computer, and operated successfully in June 1949.
He invented microprogramming while designing the control unit of the EDSAC 2, which was the first microprogrammed computer. The EDSAC 2 also used a "bit slice" design. Interchangable, replaceable tube assemblies were used for each bit of the processor. This was extremely advanced at the time.
The next computer for his laboratory was a joint venture with Ferranti Corp, the Titan. It eventually supported an early time-sharing system and provided wider access to computing resources in the university, including time-shared graphics systems (under Charles Lang) for mechanical CAD.
A notable design feature of the Titan's operating system was that it provided controlled access based on the identity of the program, as well or instead of, the identity of the user. Its programming system also had an early version control system.
The laboratory developed most of its software in Algol, and contributed to the development of Algol 68, the most advanced standard form of Algol.
Near the end of the 1960s, Wilkes became interested in capability-based computing, and the laboratory assembled a unique computer, the Cambridge CAP. It was programmed in Algol 68c.
In 1974 Wilkes encountered a Swiss data network (at Hasler AG) that used a ring topology to allocate time on the network. The laboratory initially used a prototype to share peripherals. Eventually, commercial partnerships were formed, and similar technology became widely available in England.
In 1980 he retired from his professorships and Directorate of the laboratory and joined the central engineering staff of Digital Equipment Corp. in Maynard, Massachusetts.
In 1986 he returned to England, and became a member of Olivetti's Research Strategy Board. In 2002, Wilkes moved back to the Computer Laboratory, University of Cambridge, where he is currently an Emeritus Professor.
See also: his colleague Roger Needham.