Dr. Allan L. Scherr

A pioneer in the development of operating systems for large-scale computers and systems for defining and automating business processes, he was born in Baltimore, Maryland. He graduated in 1958 from the Baltimore Polytechnic Institute. In 1959, he pursued his education in Electrical Engineering at the Massachusetts Institute of Technology (MIT), receiving his B.Sc. and M.Sc. degrees in 1962 and his Ph.D. in Electrical Engineering in 1965; his thesis being, "An Analysis of Time-Shared Computer Systems." At MIT, he was first exposed to computers in a programming course taught by John McCarthy (the inventor of LISP) and Nat Rochester (a future IBM fellow). His concentration during his undergraduate and Master's degree studies was in computer logic design. During this period of time he also worked as a cooperative student with IBM doing logic design with their advanced technology groups; which work resulted in several patented inventions. In 1963 he was part of the original group of graduate students at MIT's Project MAC (which later became the laboratory for Computer Science). His Ph.D. research involved measuring and modeling the performance of the world's first general-purpose time-sharing system, CTSS. This work, still referenced in system performance analysis literature, earned him the ACM's Grace Murray Hopper Award in 1975, and was published as a research monograph by the MIT Press. He joined IBM as a staff engineer in 1965 and worked on the architecture for what became the IBM System/370 line. His work in simulating program-addressing patterns and memory allocation was instrumental in establishing IBM's virtual storage architecture and the direction for memory allocation in OS/360. He went on to lead a series of studies in 1967 to create a time-sharing system strategy for IBM. His proposal for a general-purpose time-sharing system was accepted and he became the Manager of the group that designed and led the implementation of TSO (Time Sharing Option), today's most widely used time-sharing system. In 1971 he participated in the IBM task force that proposed the creation of MVS, a multiple virtual storage, multiple coupled processor, high-visibility operating system for IBM's large computer systems. He became the overall manager for the project until its shipment in 1974. This was the largest single software release ever produced, consisting of nearly two million lines of new and changed code on a base of over three million lines. he took over the management of the development of a new operating system for distributed processing (DPPX) for IBM's minicomputer line, the 8100 in 1977 and in 1980, after successfully shipping the first release of DPPX, he was named a Director of Programming in the Systems Communications Division, supervising the development of IBM's networking software (VTAM, TC, NCP) and its premier transaction processing system (CICS). After a two-year assignment on the corporate engineering and programming staff, in 1983 he directed the early design work for what later evolved into AS/400, IBM's midrange system line. During this period he also developed an approach to achieving unprecedented levels of productivity from engineering and programming teams. In 1986 he moved to IBM's Application Solutions Line of Business, where he directed engineering and programming groups developing products for specific application by IBM customers in various industries. In 1989 he directed the technical staff overseeing development for the entire line of business and coordinated the creation of an overall application architecture. He represented the application layer of software in the creation of IBM's System Application Architecture and was instrumental in setting its distributed-processing and intelligent-workstation direction. In 1991 he became Vice President of Technology in the IBM Consulting Group that was responsible for providing leading-edge tools and methodologies for consultants in advising clients on information technology strategy, business process reengineering, and quality. Since then he has led a research effort in the definition, automation, and management of business processes. His group has developed a new approach to defining business processes that promises to revolutionize the way computers are used in business. He has received numerous honors and awards; namely, the IBM Outstanding Contribution Award for TSO Design and Development in 1971; the ACM Grace Murray Hopper Award for pioneering work in performance analysis (of time-sharing systems) in 1975. He was named an IEEE fellow in 1983; and an IBM Fellow for his development of IBM's general-purpose time-sharing system, the large-scale operating system, MVS, and software support for distributed processing and communications networks in 1984. Among the many publications he has authored or co-authored are: "The Evolution of the MVS Operating System," with Larkin Auslander, IBM J. Research & Development, Vol. 25, No. 5, 1981, pp. 471-482; ?An Analysis of Time-Shared Computer Systems?, MIT Press, Cambridge, Mass., 1966; "Time-Sharing Measurement," Datamation, Apr. 1966; "Functional Structure of IBM Virtual Storage Operating Systems; Part II: OS/VS2-2 Concepts and Philosophies," IBM Systems J, Vol. 12, No. 4, 1973, pp. 382-400; "Distributed Data Processing," IBM Systems J., Vol. 17, No. 4, 1978, pp. 324-343; "A Perspective on Communications and Computing," IBM Systems J, Vol. 22, Nos. 1 & 2, 1983, pp. 5-9; "Structures for Networks of Systems," IBM Systems J., Vol. 26, No. 1, 1987, pp. 4-12; "SAA Distributed Processing," IBM Systems J., Vol. 27, No. 3, 1988, pp. 370-383; "Managing for Breakthroughs in Productivity," Human Resource Management, Vol. 28, No. 3, 1989, pp. 403-424; and "A New Approach to Business Processes," IBM Systems J, Vol. 32, No. 1, 1993. Karen A. Frenkel profiled his work in her publication, ?Allan L. Scherr: Big Blue's Time-Sharing Pioneer,? Profiles in Computing, Comm. ACM, Vol. 30, No. 10, Oct. 1987, pp. 824-828; revealing a man at the center of large projects, managing hundreds of people scattered across the globe and fashioning programs having thousands of line of code.