Showing papers on "Overhead (computing) published in 1977"

Shifting garbage collection overhead to compile time

Abstract: This paper discusses techniques which enable automatic storage reclamation overhead to be partially shifted to compile time. The paper assumes a transaction oriented collection scheme, as proposed by Deutsch and Bobrow, the necessary features of which are summarized. Implementing the described optimizations requires global flow analysis to be performed on the source program. It is shown that at compile time certain program actions that affect the reference counts of cells can be deduced. This information is used to find actions that cancel when the code is executed and those that can be grouped to achieve improved efficiency.

Illuminated overhead advertising display

Abstract: An overhead sign having polyhedral sections tapered vertically in opposite directions, the lower section comprising downwardly converging translucent panels having picture transparencies backlighted by concentrated light from overhead fluorescent room lights while upwardly converging decorative opaque panels overhang the upper edges of the translucent panels to minimize front lighting and perceptively brighten the pictures viewed by customers as well as minimizing shadows on the display counters below the sign.

Van overhead console

Abstract: A four-piece molded console to be mounted on the interior roof and rear corners of a van. The overhead portion is a two-piece rectangular shape with a matching piece at each side of the rear to fit the rear corners of the van. The console contains numerous compartments and cutouts in which are installation speakers, mirrors, electronic equipment and controls, a clock, vents, lights and at least one openable window.

Battery powered public transport vehicle - has overhead pantographs collecting current to recharge batteries from overhead wires situated at stopping places

Abstract: The vehicle is propelled by electric motors running off a bank of accumulators (11), which are of the rapid charging type. Overhead pantographs (2, 3) at the front and rear of the vehicle collect current from overhead wires (6, 7) situated at stopping points on the vehicle route. The cables are suspended from a gantry (10) and are an earth cable (6), and a live cable (7) connected to an underground live mains cable (8). The amount of battery discharge by travelling between recharge points is a small fraction of the total charge held, and recharging occurs rapidly. The vehicle can be powered by the overhead lines directly while it sets off from the station.

An analytic model for parallel computation

Abstract: A multiprocessor (MP) system is defined to be two or more independent processing units accessing a common memory for instructions and data. The common memory may be divided into independently accessible banks. A parallel program is a program operating on an MP system which makes use of more than one processing unit to achieve its computational goals.An analytic model is constructed which exposes certain properties of parallel programs, based on certain idealized assumptions. These assumptions are that:1. the program algorithm may be divided into independent parallel-executable sections in any way desired;2. the program is compute-bound, so that input-output considerations do not affect its behavior;3. there exist linear cost functions for elapsed time and for computer time used;4. initiation and termination of each independent activity of a parallel program incurs an identical amount of overhead;5. competition for access to storage causes degradation in performance, manifested as reduced computational speed; and6. references to storage are independently and uniformly distributed.Under these assumptions, the model gives an optimal number of processing units to use for the program in order to achieve minimum cost. This optimum is dependent on the overhead, the degradation, and the ratio of the cost of elapsed time to the cost of computer time. In the limit of very large compute time, a bound for the cost ratio is established, below which the use of parallel methods is not economically feasible. This lower bound depends only on the behavior of the degradation factor. Further analysis shows that the degradation factor is linear in the number of processing units employed, inversely linear in the number of independent storage banks, and quadratic in the relative storage accessing rate.A Monte Carlo model was constructed by J. L. Rosenfeld which gave similar results for those situations where the assumptions of the two models coincided.

Fast algorithms for load-flow problems

Abstract: In a recent paper [1], a model of a parallel processing system (PAPROS) was described and associated algorithms for solving the load flow problem were developed. In order to evaluate gains brought about by using PAPROS instead of a standard mono-computer, an “overhead” factord will be defined and computed in Section 2. This computation shows that for parallel execution the Gauss- Jordan method is preferable to straightforward elimination and the reasons for this unexpected result are explained. In Section 3, two new factorization methods, the “block-elimination” and Danilewski algorithms, are discussed. The first is superior to known methods, if executed on PAPROS. In Section 4 a general method to speed up the solution of direct methods is developed. In conclusion (Section 5) the contributions of the paper are listed.

A conceptual design of a generalized database subsystem

Abstract: A database subsystem (GDS) is proposed. The main objective is to provide a basis for constructing efficient database management systems. A functional interface (GDI), defined for the subsystem, primarily manipulates a single record. Various access functions are provided in order to manipulate a wide variety of logical data structures. Mapping from a logical data structure to a physical data structure is completely maintained in the GDS. Data definition and control functions are also included in the GDI. The subsystem is implemented on a special purpose processor (GDP) which independently operates from a host processor, responsible for the rest of the database management functions. The GDP is composed of special hardware and firmware function modules, local memories and a shared memory which is used for communicating with the host processor. Parallel processing at the host-GDP level and intra-GDP level assures high performance. Furthermore, the interprocessor communication mechanism using the shared memory significantly reduces overhead time by avoiding useless data transfer operations. The subsystem can be applied to various database management system, such as a CODASYL type, a relational type, end user facilities etc.

Arrangement for applying written or drawn information onto image carriers for overhead projection

Abstract: Arrangement for applying written or drawn information onto image carriers for overhead projectors, characterized by an image carrier in film or plate form which comprises at least one colored compound, as well as a writing or drawing marker. The writing or drawing fluid of the marker contains at least one colorless compound which reacts with the colored compound to produce a color change.

Background information on high voltage fields.

Abstract: The increased demand for power has led to higher voltages for overhead transmission lines Environmentalists, governmental agencies, and some members of the scientific community have questioned if

Road rail container transfer installation - has mobile overhead gantry interposed between overhead cables and train for indirect and direct transfer

Abstract: The container transfer method has a mobile lifting device interposed between the electric overhead cables of the railway and the train. The lifting device moves on an overhead gantry supported by drivable wheels on rails. The overhead gantry may be moved along the train or parallel to it with the lifting device remaining between the overhead cable and the train. The transfer installation includes several transfer lines on the loading ramp, thus permitting two-way transfer operations so that one item may be placed on the vacant spot in one line with another item being loaded on the same truck without any length movement of the gantry. In addition this layout permits direct transfer of containers from the railway truck to a road vehicles and vice versa.

A new methodology for computer system data gathering

Abstract: Many computer system monitoring, data gathering, and reduction efforts ignore unbiased sampling techniques. The approaches generally taken are expensive and can make no scientifically based statement about the accuracy of the data gathered or consequent data reduction. The approach outlined in this paper attempts to correct these inadequacies by using the theory of random sampling.Several new techniques are introduced for obtaining optimal error bounds for estimates of computer system quantities obtained from random samples. A point of view is taken (boolean variable random sampling) which makes it unnecessary to have any a priori knowledge of the population parameters of the phenomena being sampled. It is expected that the techniques introduced will significantly reduce monitoring overhead for computer systems while increasing the quality of the data gathered.

Data processing unit

Abstract: PURPOSE:To reduce system overhead caused by overlapping of writing operation to the main memory unit caused by speed difference between the processing unit and main memory unit and to improve processing speed of the processing unit

Mass effective affix damper of overhead transmission line

Abstract: PURPOSE:Mapping out to simplify the design, the production, the transport and the assembling works in consideration of individual manufacturing and assembly every unit, to make the frequency characteristic into broad-band

Response method of communication control system

Abstract: PURPOSE:To eliminate extra overhead of the host computer with respect to invalid pollings by making it possible to process all negate response transmission processings with respect to invalid pollings in the communication control unit.

A few examples of how to use a symmetrical multi-micro-processor

Abstract: This paper presents an overview of an architecture of a multi-micro-processor architecture in which all processors are equivalent. The structure is characterized by a central memory and several caches, and is designed to avoid incoherent data. A fast mechanism of “subcontracting” between one processor and another is described.The execution of two simple programs under this architecture is studied. The parallel algorithms are described and compared with the original (sequential) algorithms; it is shown that the overhead remains small, few extra memories are necessary and synchronization do not slow down the execution unduly. Furthermore, the locality of the parallel algorithm is compared to that of the sequential algorithm and is found to be less good, but in a reasonable way. As well, an estimation of the cost of the coherence keeping mechanism is given using the difference between the miss ratio of the classical cache and the “coherence keeping cache”, respectively.

Data processing system

Abstract: PURPOSE:The shared bus control unit executes as proxy the interruption processing according to the type of interruption factors and reports the interruption signal to plural processing units simultaneously according to the type of interrution factors, and therefore the processing request interruption is inputted at the almost same time, so that it is possible to reduce considerably the overhead of the interruption processing.

Code error correcting method

Abstract: PURPOSE:To prevent unnecessary increase in the scale of circuit and the number of overhead bit, and obtain a practical signal-to-code error noise ratio by converting important higher number bits only among those constituting a code into an error correcting code in PCM communication.

Snakeetravel preventing apparatus for an overhead travelling crane

Abstract: PURPOSE:To provide a snake-travelling preventing apparatus for an overhead travelling crane, in which a fastening member having rail holding and guiding rollers pivotally secured thereto is mounted on saddles through the intermediary of centering return springs, and one of the saddles is fixed to the main girdle while the other is mounted in the manner that is slidable in the direction of span, thereby positively preventing snaketravel of the overhead travelling crane.

Interior board for vehicles and a method of making the same

Abstract: PURPOSE:Method of mass-producing interior boards of vehicles such as door trims, ceiling boards, dash boards, overhead consoles, etc. in sufficiently rigid construction, light weight, easily and cheaply.

Estimating costs and performance of systems for machine processing of remotely sensed data

Abstract: This paper outlines a method for estimating computer processing times and costs incurred in producing information products from digital remotely sensed data. The method accounts for both computation and overhead, and may be applied to any serial computer. The method is applied to estimate the cost and computer time involved in producing Level II Land Use and Vegetative Cover Maps for a five-state midwestern region. The results show that the amount of data to be processed overloads some example computer systems, but that the processing is feasible on others.