Because of the increasing complexity and flexibility of modern highly automated production systems solutions are required for the simulation of entire production environments including all hardware and software compon...
详细信息
ISBN:
(纸本)1892512459
Because of the increasing complexity and flexibility of modern highly automated production systems solutions are required for the simulation of entire production environments including all hardware and software components inside an installation. In this regard, we introduce a unique approach to factory simulation based on hierarchically distributed model partitions and rest upon 3D graphical simulation. Thus, we are able to process models of complex environments. Using the identical interface developed for distributed computing our approach supports the necessary coupling to existing real production controllers. Moreover, the functionality of the evolved. interface between simulator's instances facilitates the connection of other simulation tools to our system. In this paper, we focus on the parallel execution of hierarchically distributedprocessing and coupling of manufacturing simulations and production control.
This paper describes the successful parallel implementation of genetic programming on a network of processing nodes using the transputer architecture. With this approach, researchers of genetic algorithms and genetic ...
详细信息
This paper describes the successful parallel implementation of genetic programming on a network of processing nodes using the transputer architecture. With this approach, researchers of genetic algorithms and genetic programming can acquire computing power that is intermediate between the power of currently available workstations and that of supercomputers at intermediate cost. This approach is illustrated by a comparison of the computational effort required to solve a benchmark problem. Because of the decoupled character of genetic programming, our approach achieved a nearly linear speed up from parallelization. In addition, for the best choice of parameters tested, the use of subpopulations delivered a super-linear speed-up in terms of the ability of the algorithm to solve the problem. Several examples are also presented where the parallel genetic programming system evolved solutions that are competitive with human performance. (C) 1998 Published by Elsevier Science Inc. All rights reserved.
Control of large flexible structures such as segmented telescopes requires high computational power to achieve the desired performance in real-time. Through decentralized control, the overall structure can be controll...
详细信息
ISBN:
(纸本)1892512459
Control of large flexible structures such as segmented telescopes requires high computational power to achieve the desired performance in real-time. Through decentralized control, the overall structure can be controlled by multiple lower-order local controllers, leading to reduced computational complexities. In addition, the decentralized approach can both simplify the development of parallel programs and facilitate the study of fault-tolerant control systems. This paper describes the real-time embedded computing system to perform the control algorithms on a segmented reflector telescope test-bed. It also demonstrates the inherent link between decentralized control and concurrent processing by its parallel implementation.
The linear complementarity problem LCP(A, q) which consists of finding a vector z is an element of R-n such that Az + q greater than or equal to 0, z greater than or equal to 0, z(T) (Az + q) = 0, where A is an elemen...
详细信息
ISBN:
(纸本)1892512459
The linear complementarity problem LCP(A, q) which consists of finding a vector z is an element of R-n such that Az + q greater than or equal to 0, z greater than or equal to 0, z(T) (Az + q) = 0, where A is an element of R-nxn and q is an element of R-n are a given real matrix and an real vector, respectively. This paper proposes an O(n(2)) parallel algorithm (or O(n) parallel algorithm) by using n processors (or n(2) processors) for solving the LCP(A, q) with A is an M-matrix.
Let S : [0, 1](2) --> [0, 1](2) be a nonsingular dynamical system and let P : L-1([0, 1](2)) --> L-1([0, 1](2)) be the corresponding Frobenius-Perron operator. In this paper we study our parallel algorithm to im...
详细信息
ISBN:
(纸本)1892512459
Let S : [0, 1](2) --> [0, 1](2) be a nonsingular dynamical system and let P : L-1([0, 1](2)) --> L-1([0, 1](2)) be the corresponding Frobenius-Perron operator. In this paper we study our parallel algorithm to implement Ulam's method for computing the fixed density of P with the quasi Monte Carlo (QMC) approach. Numerical results with the efficiency of parallel algorithm are presented.
parallel and efficient sorting of large amount of data is very important for real life applications. In this project, a parallel sorting algorithm is designed and implemented on a cluster of Linux workstations using M...
详细信息
ISBN:
(纸本)1892512459
parallel and efficient sorting of large amount of data is very important for real life applications. In this project, a parallel sorting algorithm is designed and implemented on a cluster of Linux workstations using MPI and PVFS as a parallel file system. Two versions of the sort are studied, including a sort in which all data fits into the collective memory of the cluster, and a version in which the data does not fit into the collective memory of the cluster. Efficiency of the implemented algorithms are analyzed. Two implementation issues are addressed: the use of non-blocking sends and receives in places of MPI library calls for collective operations, and the number of PVFS files used as temporary storage. Results show that the performance of out-of-core sorts can be improved significantly with certain implementation optimizations.
This paper presents a self-stabilizing algorithm for mutual exclusion in asynchronous networks. The algorithm is token based and a binary tree network topology is used The algorithm is resilient to transient faults an...
详细信息
ISBN:
(纸本)1892512459
This paper presents a self-stabilizing algorithm for mutual exclusion in asynchronous networks. The algorithm is token based and a binary tree network topology is used The algorithm is resilient to transient faults and does not require initialization. It has been proved that the algorithm is correct and requires O(n) moves, where n is the number of nodes in the system.
A parallel volume rendering system for unstructured grid volume data is proposed in this paper. By implementing the mechanism for dynamic load balancing into the system, the authors solve the issue of load imbalance d...
详细信息
ISBN:
(纸本)1932415262
A parallel volume rendering system for unstructured grid volume data is proposed in this paper. By implementing the mechanism for dynamic load balancing into the system, the authors solve the issue of load imbalance due to the view dependency and run-time features of early ray termination. An experimental implementation of the system achieved a 4.32-times performance improvement.
parallel normalized preconditioned conjugate gradient type methods based on normalized approximate finite element inverse matrix techniques are investigated for the efficient solution of sparse linear systems. Applica...
详细信息
ISBN:
(纸本)1932415262
parallel normalized preconditioned conjugate gradient type methods based on normalized approximate finite element inverse matrix techniques are investigated for the efficient solution of sparse linear systems. Application of the proposed methods on a three dimensional boundary value problems is discussed and numerical results are given. The parallel implementation of the normalized preconditioned conjugate gradient-type I methods is also discussed and speedups are presented.
The use of hierarchical clustering in distributed systems shifts the but-den of workflow management from a central server to work managers participating in the system, as well as necessitating a node profiling mechani...
详细信息
ISBN:
(纸本)1932415262
The use of hierarchical clustering in distributed systems shifts the but-den of workflow management from a central server to work managers participating in the system, as well as necessitating a node profiling mechanism. Examination of this shift illuminates the need for alternative approaches to managing work unit distribution and result collection. This paper describes our intended initial approach to work assignment along with a simple formal model for workflow in the system and some of the issues that the system will need to address as it progresses.
暂无评论