Corpus ID: 195607370

Validity of the Single Processor Approach to Achieving Large Scale Computing Capabilities

  title={Validity of the Single Processor Approach to Achieving Large Scale Computing Capabilities},
  author={G. Amdhal},
An instrument for facilitating the calculation of equivalent values includes a plate bearing symbols representing units and dimensions, the plate having a window in which a movable pointer is located. The pointer is movable from a first position representing a first set of units and dimensions to a second position representing an equivalent set of units and dimensions. 
The Myth of Performance for Parallel Machines
It has been a decade since the first CRAY-1 computer was delivered to Los Alamos National Laboratories in the U.S. in 1976 and the fact that IBM introduced their own vector unit for the Sierra series in 1985 is the ultimate acknowledgement that vector processing is an accepted standard mode of operation in the conventional camp. Expand
Architecture for large computer systems
As the computer field reaches maturity, users select their equipment not because of internal structure of features, but by its overall economy; i.e., performance per unit cost. Here, performanceExpand
Advanced Architecture Computers
We describe the characteristics of several recent computers that employ vectorization or parallelism to achieve high performance in floating-point calculations. We consider both top-of-the-rangeExpand
Short communication acritical for parallel processors
Values of acritird are given for a number of contemporary high performance computing machines. Expand
Associative and Parallel Processors
This paper covers the topics of system categorizations, applications, main tradeoff issues, historically important architectures, and the architectures of systems that are currently available on the subject of parallel and associative processors. Expand
Three-dimensional magnetic field computation on a distributed memory parallel processor
The author reports on an implementation of a finite-element code for nonlinear three-dimensional low-frequency magnetic field calculation on Intel's iPSC/2. The approach used exploits the naturalExpand
Vector System Performance of the IBM 3090
The paper has two parts, the first presenting factors affecting performance measurement of the Vector Facility and the criteria for its design and the second using use of the 3090 storage hierarchy to support the vector processing implementation. Expand
Observations on high-performance machines
  • D. Senzig
  • Computer Science
  • AFIPS '67 (Fall)
  • 1967
The high speed computer area seems to be dominated by a continued reduction in the price of computer switching circuits and the approach of these circuits to speeds at which the velocity of light becomes an important factor, so simultaneousity becomes more and more essential if computer performance is to continue to increase. Expand
Senior Project : Parallel Programming
After years of technological advances the speed of single processors are beginning to meet their physical limitations. Thus, parallel programming has become an increasingly important tool inExpand