Showing papers on "Benchmark (computing) published in 1981"
TL;DR: This paper uses DMIN, an optimal dynamic aliocation algorithm, to compute the minimum space-time cost achievable for some benchmark program runs, and compares it with that from MIN, an ideal static allocation algorithm, VMIN, a optimal variable space algorithm, and two heuristic dynamic allocation algorithms.
Abstract: In this paper we compare the performance of virtual memory allocation algorithms. The primary measure of performance is the space-time product of primary memory occupancy, or space-time cost, used by a program during its execution. Using DMIN, an optimal dynamic aliocation algorithm, we compute the minimum space-time cost achievable for some benchmark program runs. We compare the DMIN space-time cost with the space-time cost from: MIN, an optimal static allocation algorithm, VMIN, an optimal variable space algorithm, and two heuristic dynamic allocation algorithms. the page fault frequency algorithm and the damped working set algorithm.
8 citations
TL;DR: In this paper, an expansion for the combining coefficients of singular eigenfunctions is developed that yields exact elementary plane-wave solutions that can be evaluated very accurately in terms of exponential integral functions and linear combinations of these solutions are automatically generated which approximate, in the least squares sense, a user specified incident flux distribution on the boundary of a multidimensional benchmark cell domain.
4 citations
01 Jan 1981
TL;DR: It is concluded that the speed of the Cyber-205 for memory-to-memory operations on vectors stored in consecutive locations is considerably faster than that of the Cray-1, but the overall performance of the machine is not quite equal to that ofThe Cray for tasks of interest to the Laboratory as represented by the benchmark set.
Abstract: This report compares the performance of Control Data Corporation's newest supercomputer, the Cyber-205, with the Cray Research, Inc. Cray-1, currently the Laboratory's largest mainframe. The rationale of our benchmarking effort is discussed. Results are presented of tests to determine the speed of basic arithmetic operations, of runs using our standard benchmark programs, and of runs using three codes that have been optimized for both machines: a linear system solver, a model hydrodynamics code, and parts of a plasma simulation code. It is concluded that the speed of the Cyber-205 for memory-to-memory operations on vectors stored in consecutive locations is considerably faster than that of the Cray-1. However, the overall performance of the machine is not quite equal to that of the Cray for tasks of interest to the Laboratory as represented by our benchmark set.
4 citations
Journal Article•
TL;DR: It is found that two modifications on core-reflector boundary conditions and Brrensen's model constants may improve computational accuracy of power distribution calculation.
01 Jan 1981
TL;DR: "Benchmark" programs are being developed to establish certain figures of merit to determine execution speed and program administrative costs and a detailed analysis of these results indicates strengths and weaknesses of benchmarking in a static and dynamic execution.
Abstract: The University of Nevada-Reno is conducting research relative to comparing several microprocessors in an automatic testing environment, "Benchmark" programs are being developed to establish certain figures of merit to determine execution speed and program administrative costs. Furthermore, a detailed analysis of these results indicates strengths and weaknesses of benchmarking in a static and dynamic execution. Three microprocessors have been considered. The Intel 8086 16-bit machine, the Motorola 6801 8-bit machine and a Motorola 6800 with an AMD-9511 arithmetic unit. Currently a Motorola 68000 and a TI 990 are under investigation. Emphasis is placed on arithmetic and logical operations with additional input/output instructions included in the execution time.