- The term architecture is used here to describe the attributes of a system as seen by the programmer, i.e., the conceptual structure and functional behavior, as distinct from the organization of the data flow and controls, the logical design, and the physical implementation. i. Additional details concerning the architecture,
- [The architecture specification covers] all functions of the machine that are observable by the program.
- G.A. Blaauw, F.P. Brooks, "The Structure of System/360", IBM Systems Journal, Vol.3, No.2, p. 119-135, 1964; as cited in: Vojin G. Oklobdzija (1999) "Reduced instruction set computing"
- The architecture of a system can be defined as the functional appearance of the system to the user.
- Blaauw (1972) cited in: Gerritt A Blaauw (1976) Digital system implementation. p. 6
7000QX Committee Interim Report (1958)
G.A. Blaauw & H.G. Kolsky (1958) 7000QX Committee Interim Report. Project 7000. Product Development, Laboratory Poughkeepsie, N.Y. December 31, 1958
- The purpose of the Committee was to study and report upon the desirability and characteristics of another computer system based on Stretch technology but having a lower cost and broader market than the 7000 Sigma system. The Committee was instructed to keep an open mind in initially examining various machine possibilities both from the engineering and marketing points of view...
- p. 1
- 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.
- p. 3
- 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.
- p. 3: About reprogramming applications
Computer architecture (1972)
Gerrit Blaauw (1972) "Computer Architecture". Elektron. Rechenanl. Vol 14 (2) p. 154-159
- In computer design three levels can be distinguished: architecture, implementation and realisation; for the ﬁrst of them, the following working deﬁnition is given: The architecture of a system can be defined as the functional appearance of the system to the user, its phenomenology.
- Although the term architecture was introduced only ten years ago in computer technology (Buchholz), the concept of architecture is as old as the use of mechanism by man. When a child is taught to look at a clock, it is taught the architecture of the clock. It is told to observe the position of the short and the long hand and to relate these to the hours and the minutes. Once it can distinguish the architecture from the visual appearance, it can tell time as easily from a wrist watch as from the clock on the church tower.
- The inner structure of a system is not considered by the architecture: we do not need to know what makes the clock tick, to know what time it is. This inner structure, considered from a logical point of view, will be called the implementation, and its physical embodiment the realisation.
- p. 154
- 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.
- p. 154
- 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.
- p. 155
Specification of Digital Systems (1978)
G.A. Blaauw (1978). "Specification of Digital Systems". Proc. Seminar in Digital Systems Design (Lecture Notes). University of New Castle upon Tyne. , Engeland
- 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.
- p. 29
- By architecture I mean 'appearance to the user' - it is the functional specification of the system (its behavioural appearance). By implementation I mean 'internal logical organisation which performs the functions specified by the architecture' and by realisation I mean 'the physical components in which the logical organisation is embodied'.
- p. 29
- A hardware design language should be (i) sufficiently high level, (ii) conversational, (iii) general purpose, and (iv) structured:
- (i) High Level: The language should be able to express easily , and directly the desired function . Also, it should avoid suggesting an implementation (fo r example, by introducing unnecessary concepts). It should avoid involved or clever language constructs and it should be understandable to the users (that is, readable, pronouncable and not cryptic) .
- (ii) Conversational: A language that is available conversationally aids the description by the elimination of errors through syntax checks. Also, the description is executable and thus demonstrates the architecture. It is of course not possible to verify the architecture - there is nothing to compare it with. What is possible is to check the implementation against the architecture . By selectively substituting implementation for architecture one can perform efficient simulation experiments.
- (iii) General purpose: When the language is general purpose the same language can be used for all levels of design (architecture, implementation, realisation) . Similarly, design tools necessary for testing, collecting performance statistics, editors, tracers etc . can also be built using the same language . Thus a general purpose l anguage represents a desirable economy of thought for the designer.
- (iv) Structured: A language that exhibits good structure permits the designer to show the structure of his design . It allows the designer to develop his specifications in a top-down way so that design details can be developed and expressed gradually.
- p. 30
Computer architecture: concepts and evolution (1997)
Gerrit A. Blaauw, Frederick Phillips Brooks (1997) Computer architecture: concepts and evolution. Addison-Wesley
- 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 book's focus on "architecture" issues complements Blaauw's 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.
- Blaauw and Brooks first develop a conceptual framework for understanding computer architecture. They then describe not only what present architectural practice is, but how it came to be so. A major theme is the early divergence and the later reconvergence of computer architectures. They examine both innovations that survived and became part of the standard computer, and the many ideas that were explored in real machines but did not survive. In describing the discards, they also address "why" these ideas did not make it
About Gerrit Blaauw
- As usual the audience consisted mainly of professors of computing science; this time the speakers were mainly specialists in logic design: for many in the audience the exposure was a shock. At the level of component technology the change over the last fifteen years has been drastic: what used to be expressed in milliseconds is expressed in microseconds now, what used to be expressed in kilobucks is now expressed in dimes and quarters. This change has been so drastic that it is well-known. Much less known is that at the next levels, viz. of circuit design and logic design, the attention of the designers has been so fully usurped by the obligation to adapt to the ever changing technology, that at those levels design methodology has had no chance to mature from craft to scientific discipline. This is in sharp contrast to the developments in programming methodology, where during that period of fifteen years a fairly stable "base" could be enjoyed. Having witnessed that development in programming methodology at close quarters, I was overcome by the feeling of being exposed to the result of fifteen years of intellectual stagnation, and it was during Blaauw's lecture on the first afternoon that I asked my right-hand neighbour "Close your eyes, forget how you came here and guess in which year you are living."; without hesitation he came up with exactly the same year I had in mind: 1962.
- Edsger W. Dijkstra (1977-82) Selected Writings on Computing: A Personal Perspective, Springer-Verlag, 1982. ISBN 0–387–90652–5. Chapter . Tripreport E.W.Dijkstra, Newcastle-upon-Tyne, 5 - 10 Sept. 1977. About the 1977 "Joint International Seminar on the Teaching of Computing Science", in Newcastle-upon-Tyne, England.
- The term Computer Architecture was first defined in the paper by Amdahl, Blaauw and Brooks of International Business Machines (IBM) Corporation announcing IBM System/360 computer family on April 7, 1964. On that day IBM Corporation introduced, in the words of IBM spokesman, "the most important product announcement that this corporation has made in its history".
- Computer architecture was defined as the attributes of a computer seen by the machine language programmer as described in the Principles of Operation. IBM referred to the Principles of Operation as a definition of the machine which enables machine language programmer to write functionally correct, time independent programs that would run across a number of implementations of that particular architecture.
- The architecture specification covers: all functions of the machine that are observable by the program. On the other hand Principles of Operation. are used to define the functions that the implementation should provide. In order to be functionally correct it is necessary that the implementation conforms to the Principles of Operation...
- Vojin G. Oklobdzija (1999) "Reduced instruction set computing"
- Practical knowledge of modularity has come largely from the computer industry. The term architecture was first used in connection with computers by the designers of the System/360: Gene M. Amdahl, Gerrit A. Blaauw, and Frederick P. Brooks.
- James H. Gilmore, B. Joseph Pine (2000) Markets of One: Creating Customer-unique Value Through Mass. p. 51
- Blaauw... joined the IBM research lab at Poughkeepsie, New York, USA. During this period Blaauw became famous for his methodological manner of building computer machines. He made a difference between architecture, implementation and realization of a machine. After a couple of years working on some different machines, one of the most famous machines built by Frederick Brooks, Gerrit Blaauw and Gene Amdahl was the IBM System/360, which was introduced in 1964. IBM Board Chairman Thomas Watson, Jr. called the event the most important product announcement in the company's history.
- Gerrit Blaauw, unsung heroes in Dutch computing history, 2007
- Gerrit Blaauw, unsung heroes in Dutch computing history.
- Diesrede Blaauw: Beschrijven en begrijpen 26 November 1976 (in Dutch)
- Data processing system programmed by instruction and associated control words including word address modification. United States Patent 3202971