the aim of this paper is to present the results of our survey on the applications and problems arising from parallelprocessing in the area of Airlines Reservations Systems. We used the Reservation System of Olympic A...
详细信息
the aim of this paper is to present the results of our survey on the applications and problems arising from parallelprocessing in the area of Airlines Reservations Systems. We used the Reservation System of Olympic Airways as our basic model. We shall describe the systems, the problems which arise and finally we shall give our propositions and solutions.
this contribution describes a new class of arithmetic architectures for Galois fields GF(2k). the main applications of the architecture are public-key systems which are based on the discrete logarithm problem for elli...
详细信息
A distributed optimization method for the minimization of the total resource of a system with discrete elements is presented, and a theoretical and experimental investigations are carried out in this paper. the distri...
详细信息
ISBN:
(纸本)1853124583
A distributed optimization method for the minimization of the total resource of a system with discrete elements is presented, and a theoretical and experimental investigations are carried out in this paper. the distributed optimization algorithm consists of two processes, namely the resource reduction process and the resource addition process. In the former process, each element discards its critical resource margin with respect to the global and local constraints, while in the latter process, a small amount of resource is added to each element. the proposed method is successively applied for optimizing truss structures, and the method is found to be very robust and suitable for parallelprocessing.
this paper proposes a parallel algorithm based upon dynamic programming, for velocity field estimation. It has O(N) complexity, N being the number of element involved in the process. this low complexity is very intere...
详细信息
this paper proposes a parallel algorithm based upon dynamic programming, for velocity field estimation. It has O(N) complexity, N being the number of element involved in the process. this low complexity is very interesting for many real time applications (autonomous robot navigation, trajectory matching etc..). the algorithm has been implemented on the CM-5, therefore functional specifications of our parallel dynamic programming circuit have been validated.
the group-theoretic approach to the decomposition of the basic operations of the digital signal processing (DSP) such as discrete Fourier transform (DFT) and convolution is proposed. the distinctive feature of the app...
详细信息
the most encountered problem in modeling curves and surfaces by polynomial or Bezier representation is to find a new model for only a portion of the data that we want to manipulate without using them ah. One solution ...
详细信息
the most encountered problem in modeling curves and surfaces by polynomial or Bezier representation is to find a new model for only a portion of the data that we want to manipulate without using them ah. One solution is to manipulate directly the portion of the curve or surface generated from the model. Another solution is to generate a new model which redefines this portion of the curve or surface, A new method is described for constructing a polynomial or a Bezier piecewise model from the original one. the constructed piecewise model generates the same portion of curve as the original one. It has the same order of the original model and satisfies at the edges all order continuities. the obtained model is given in a polynomial or a Bezier representation. the piecewise representation will permit carrying out tasks such as piecewise modeling, piecewise reduction, parallelprocessing and uniform representation. Our new method is not complicated because it is always given as explicit expressions in terms of control points or polynomial coefficients respectively. (C) 1997 Elsevier Science B.V.
Process technology advances tell us that the one-million gate Field-Programmable Gate Array (FPGA) will soon be here, and larger devices shortly after that. We feel that current architectures will not extend directly ...
详细信息
ISBN:
(纸本)9780897918015
Process technology advances tell us that the one-million gate Field-Programmable Gate Array (FPGA) will soon be here, and larger devices shortly after that. We feel that current architectures will not extend directly to this scale because: they do not handle routing delays effectively;they require excessive compile/place/route times;and because they do not exploit new opportunities are presented by the increase in available transistors and wiring. In this paper we describe several challenges that will need to be solved for these large-scale FPGAs to realize their full potential.
Traditionally, parallel programs were the reserve of large corporations or generously sponsored research institutions, with more modestly sized organisations not able to afford the speed up offered by parallel archite...
详细信息
ISBN:
(纸本)3540628681
Traditionally, parallel programs were the reserve of large corporations or generously sponsored research institutions, with more modestly sized organisations not able to afford the speed up offered by parallelarchitectures. However, withthe development with such software as parallel Virtual Machine (PVM) which permits a group of (possibly heterogenous) machines, to work as though they were nodes in a single parallel machine, parallelprocessing is now affordable by practically every institution. Most of the companies that could most benefit from systems like those described above, such as banks, insurance companies, those involved in simulation or engineering already have serial code written to carry out their tasks. To make the transition from serial to parallel, these companies are faced with a decision to either write the code from scratch - clearly an impractical solution, or to somehow convert their code from serial to parallel. Such a conversion is possible, but, as no automatic technique currently exists, is fraught with difficulties, requires much expertise in the area, and tends to be something of a hit and miss affair, withthe quality of the final parallel program depending very much on the ability of the programmer. there are few automatic techniques for parallelizing serial, and existing methods generally consist of a set of interactive tools to help programmers, or to identify common pieces of code that can easily be parallelized, e.g. matrix multiplication etc. the difficulty in autoparallelization, and parallelization in general, is twofold. Firstly, the identification of which areas of the code can have one or more standard transformations applied to them, and secondly discovering the optimal order in which to apply these transformations. We describe a system, Paragen II, which automatically generates a list of transformations which, when applied to a serial program, produce a parallel version. Paragen II hybridizes Genetic Programming with existi
An emerging and rapidly expanding parallel Distributed processing technology for Signal processing is the Neural Network. Artificial Neural Networks (ANNs) have been effectively used in the solution of signal processi...
详细信息
the present paper deals withthe simulation of the nonlinear and time dependent behaviour of complex structures in engineering. To overcome the limiting factors of such computer simulations - the computer run time and...
详细信息
the present paper deals withthe simulation of the nonlinear and time dependent behaviour of complex structures in engineering. To overcome the limiting factors of such computer simulations - the computer run time and the memory requirement - we use an dynamic-explicit time integration procedure for the solution of the semi-discrete equations of motion, which is very well suited for parallelprocessing. At first we give a brief review of the theoretical background of the mechanical modelling and the dynamic-explicit technique for the solution of the semi-discrete equations of motion. then the concept of parallelisation is discussed.
暂无评论