In computer design three levels can be distinguished: architecture, im — Gerrit Blaauw

"Scientific customers are traditionally less worried about reprogramming efforts than commercial customers, since many jobs are of a research nature and will be done over from time to time anyway. This is obviously true of many small and lone shot" problems. In practice, however, there are many more machine hours spent on production-type scientific problems than on those of research-type at most scientific computing installations. These production problems can be as rigid and static as any commercial job. The scientists responsible for production work will complain about reprogramming just as violently as an accountant will under the same circumstances."

"In this remarkable book on computer design, long-known in the field and widely used in manuscript form, Gerrit A. Blaauw and Frederick P. Brooks, Jr. provide a definitive guide and reference for practicing computer architects and for students. The book complements Brooks recently updated classic, The Mythical Man-Month, focusing here on the design of "hardware" and there on "software," here on the "content" of computer architecture and there on the "process" of architecture design. The books focus on "architecture" issues complements Blaauws early work on "implementation" techniques. Having experienced most of the computer age, the authors draw heavily on their first-hand knowledge, emphasizing timeless insights and observations."

"The design of a digital system starts with the specification of the architecture of the system and continues with its implementation and its subsequent realisation... the purpose of architecture is to provide a function. Once that function is established, the purpose of implementation is to give a proper cost-performance and the purpose of realisation is to build and maintain the appropriate logical organisation."

"The result of the implementation, the logical design, is traditionally shown as a series of block diagrams. These blocks represent in effect a series of statements, Actually, a direct presentation of these statements is more suitable and, although less familiar, more easily understood. The Harvard Mark IV was to large degree designed and described by such statements, as has been the case with several subsequent developments."

"There always is an architecture, whether it is defined in advance - as with modern computers - or found out after the fact - as with many older computers. For architecture is determined by behavior, not by words. Therefore, the term architecture, which rightly implies the notion of the arch, or prime structure, should not be understood as the vague overall idea. Rather, the product of the computer architecture, the principle of operations manual, should contain all detail which the user can know, and sooner or later is bound to know."

"A study of the high-speed computer market with the intention of specifying a new computer brings forth a number of interesting observations... [An] striking feature of the market is that we seem to be close to satisfying the need for present-day uses of computers, but are standing on the threshold of a vast new area of applications. This new area can be characterized by the phrase computers which interact with the outside world. This concept is called "Integrated Data Processing", "Real-Time. Operation", "Process Control", "In-Line Operation", etc, Its characteristic feature is the ability of the computer to accept and send information directly to other devices. The use of computers in this fashion is being developed in the aircraft and missile industries . It is important to note that both scientific and commercial applications are going in this direction."