Showing papers on "Benchmark (computing) published in 1975"

Static and Dynamic Characteristics of XPL Programs

Abstract: One use of performance measurement techniques is in the study of operational characteristics of programs written in high-level programming languages. Information derived from such studies can be used to construct benchmark programs and synthetic workloads,1,2detect inefficiencies in programming language implementation, and suggest possible improvements in the design of computers.3,9,10Our main interest is in the latter area: the discovery of primitive operations, implied by the semantics of a programming language, that can be added to the firmware or hardware of a computer to improve overall system performance. These computer architecture optimization techniques have been applied in several studies3,9and have been used commercially to design efficient pseudo machines for the Burroughs B1700.10,12

Seeking configurational optimization in computer systems

Abstract: This paper develops analytic techniques for achieving macro-level configurational optimization of computer systems. The computer is viewed as a network of queues with the processors, main memory, auxiliary memory and the transfer channels being the servers. The objective function to be minimized is the total system cost divided by the probability of the CPU busy time for a given user benchmark. A queueing model is developed which evaluates this probability. The scope of optimization includes the determination of the parameters related to component selection and the determination of management oriented variables like degree of multiprogramming, main memory space allotted to each program. The optimization algorithm is a modified 'Direct Search Method'.

Benchmark construction

Abstract: A partially automated method of generating benchmarks for comparison of EXEC 8 with other systems has been developed as one step in preparation for choosing a new computer at the University of Trondheim.

Computer codes and data available from the Radiation Shielding Information Center

Abstract: The Radiation Shielding Information Center (RSIC) is a technical institute serving the international shielding community. It acquires, selects, stores, retrieves, evaluates, analyzes, synthesizes, and disseminates information on shielding and ionizing radiation transport. The major activities include: (1) operating a computer-based information system and answering inquiries on radiation analysis, (2) collecting, checking out, packaging, and distributing large computer codes and evaluated and processed data libraries. The data packages include multigroup coupled neutron-gamma-ray cross sections and kerma coefficients, other nuclear data, and radiation transport benchmark problem results.

A Comparison of Two Microprogrammable Processors: PDP-11/40E and MLP- 900

Abstract: : A study was undertaken to evaluate the capabilities of two microprogrammable processors: the MLP-900, a vertically-encoded 36-bit machine at the Information Sciences Institute and available over the ARPA Network; and the PDP-11/40E, a horizontally-encoded 16-bit microprocessor at Carnegie-Mellon University. The paper presents a description of the two machines, and compares their performance on a number of benchmark programs (including an emulator for the NOVA computer). In addition, the machines are compared along dimensions of two-way conditional branch costs, basic architecture, and difficulty of programming. The PDP-11/40E performed between 10% and 25% faster on all the benchmarks except the multi-word integer multiply, where the MLP-900 was four times faster (because of its wider data path).

A comparative evaluation of the cost effectiveness of computer systems (Paper Session)

Abstract: Over the last decade, continuing advancement in semiconductor technology has resulted in significant improvement in the cost-performance ratio of microcomputer processors and the consequent proliferation of minicomputers in evidence today in an ever-widening sphere of application. Figure 1 shows the minicomputer processor pricing trends over a decade [cf. 1]. Large computers exhibit an economy of scale, evidenced by the IBM 360 series which spans two orders of mangitude in processing power [cf.2].This paper attempts to compare the cost and performance of currently available computers. The cost of computer systems can vary widely with configurations. Typical small, medium and large configurations are defined in the GML Corporation's Computer Characteristics Review [cf. 3]. It is very difficult to characterize the performance of a wide variety of computer architectures by a single number. Meaningful comparisons can be obtained by comparing the execution times for a benchmark program that typifies the problem environment. For the purpose of a general study, a simpler performance index has to be used. The register-memory add time is widely accepted for such comparisons [3]. A comparison of this performance index with some benchmark results shows it to be a reasonable measure for high level comparisons.