Computer System Architecture Research Articles

New directions in computer systems architecture

In spite of major technological developments over a thirty-year span, the architecture of computer systems has remained surprisingly stable under continued pressures for change. A number of limited architectural responses to such pressures (system improvements) are reviewed in support of this thesis. Recently, however, a maturing appreciation for the potential benefits of functional programming languages for use as a base for computer applications, juxtaposed with the VLSI opportunity, has resulted in several proposals for innovative architect area. In principle, these proposals offer attractive alternatives to conventional systems, and may well become practical alternatives as well. The relative advantages of functional languages and functional machine hosts are reviewed, and two of the more interesting proposals are singled out for extended discussion and comparison.

Book Review: The Architecture of Small Computer Systems

Book Review: The Architecture of Small Computer Systems

Microprocessors and software design tools

Software development aids available to the programmer are introduced. Application programs can be broken down into almost independent software modules. Each module can be independently developed and its interface with other modules defined. Microprocessor development systems, software components and future software are discussed. It seems that changes in computer system architecture are inevitable as LSI technology spreads computing, but this will also lead to better, more reliable software development tools.

The Architecture of Small Computer Systems

The Architecture of Small Computer Systems

Computer system architecture for laboratory automation

This paper describes the various approaches that may be taken to provide computing resources for laboratory automation. Three distinct approaches are identified, the single dedicated small computer, shared use of a larger computer, and a distributed approach in which resources are provided by a number of computers, linked together, and working in some cooperative way. The significance of the microprocessor in laboratory automation is discussed, and it is shown that it is not simply a cheap replacement of the minicomputer.

Review of Computer system architecture by M. Morris Mano. Prentice-Hall 1976.

Review of Computer system architecture by M. Morris Mano. Prentice-Hall 1976.

The concept of attached processing and attached resource computer systems

Attached Processing, an innovative concept in business data processing demanding novel computer systems architecture, divides computing into three broadly specialized functions: applications processing, data storage and management, and communications. In its realization, the Attached Resource Computer, communications are handled by combining hardware and software in a high-speed, short-range interprocessor bus managed by bipolar microprocessors; data storage and management is the province of dedicated small processors attached to the interprocessor bus through the bipolar microprocessors; applications processing is done in other small processors similarly attached but entirely unburdened by communication processing requirements. The system is capable of indefinite extension and essentially unlimited throughput at very modest cost.

A method of programming measurement and estimation

Improvements in programming technology have paralleled improvements in computing system architecture and materials. Along with increasing knowledge of the system and program development processes, there has been some notable research into programming project measurement, estimation, and planning. Discussed is a method of programming project productivity estimation. Also presented are preliminary results of research into methods of measuring and estimating programming project duration, staff size, and computer cost.

Function distribution in computer system architectures

The levelwise structuring and complexity of a computer system is presented informally and as a general model based upon the notion of abstract machines (processors), processes and interpreters. The important domains of the computer architect are considered along with historical perspectives of certain stimulae and decisions that have affected the distribution of functions amongst the various levels of computer system implementations.

A multi-microprocessor computer system architecture

The development of microprocessors has suggested the design of distributed processing and multiprocessing computer architectures. A computer system design incorporating these ideas is proposed, along with its impact on memory management and process control aspects of the system's operating system. The key design feature is to identify system processes with microprocessors and interconnect them in a hierarchy constructed to minimize intercommunication requirements.

The Impact of Electronic Disks on System Architecture

The availability of mass storage devices with access times several orders of magnitude less than rotating-device memories will have varying impacts on computer system architectures. These effects will range from simple changes such as direct substitution for disks, where significant productivity increases can be easily achieved, through more radical changes where both hardware and software are changed in structure to exploit the potential of these new devices. Such changes include architectures for transaction-based computing, real-time computing with nonresident programs, and more flexible structures for time-shared computers.

Innovations in heterogeneous and homogeneous distributed-function architectures

The character, scope, and economics of computer systems have dynamically evolved into a wide variety of architectures throughout the history of computing. The present evolutionary movement of computer system architecture development is a rapid growth toward functionally distributing computer power. It therefore seems appropriate to review. recent developments in distributed-function architecture.

Architectural assistance to software debugging aids

It has been observed [1] that 45 to 50% of programming effort is spent in debugging, checkout and testing, yet the architecture of most modern computer systems does little if anything to facilitate ease of debugging. In most batch systems the programmer is sufficiently removed from the execution of his program as to be severely handicapped in diagnosing errors. There is only so much information that can be easily obtained from a voluminous coredump, for instance. Even programmers on large timesharing systems have available at most an interactive software debugging package which operates through a combination of insertions and replacements of object code and interpretation (rather than execution) of machine code. This can get to be quite inefficient when carried to the extreme and often is useful only if the program has been processed by a special compiler.

B72-3 Computer Structures, Readings and Examples

This book represents a major contribution to the study of the architecture of computer systems. It will be, I believe, must reading for any serious student of the field.

VIPER - A Tool Concept To Control Parallel CFD Applications

Typical applications in the fields of the so-called Gmrzcf CWW#e5 need massive parallel computer system architectures. Tools like parallel debuggers, performance analysers and visualizers help the code designer to develop efficient parallel algorithms. Such tools merely support the development cycle. But the engineer who makes use of parallel high-performance computing applications, e.g. numerical simulation algorithms in Computational Fluid Dynamics (CFD), must be supported in his engineering work by another kind of tool. A user tool for the production cycle is required because conventional suggestions rely on serial batch-oriented procedures which have significant drawbacks, and therefore, they are not suitable to make use in nowadays High-Performance Computing (HPC). This paper presents the principal idea of the VIPER tool concept which enables the engineer who is sitting at his graphic workstation to interact with the parallel simulation algorithm which is running on a remote parallel computer system. Parameters of theoretical model and of the numerical method are considered as objects and available for Wme observation and Wme modification via the graphical user tool interface. First performance results are shown with the VIPER tool prototype implementation applied on the parallel Navier-Stokes solver PARNSFLEX which was running on an Intel Paragon XP/S multiprocessor system.