Wyniki wyszukiwania - BazTech

1

Analiza wydajności przetwarzania danych w chmurze obliczeniowej Windows Azure

Frączek J., Zając Ł.

Studia Informatica

|

2013

|

Vol. 34, nr 2A

97--112

PL

W artykule przedstawiono wyniki badań wydajnościowych dla rozwiązania problemu komiwojażera z wykorzystaniem zrównoleglonej wersji algorytmu genetycznego w środowisku wielordzeniowego serwera lokalnego oraz chmury obliczeniowej Windows Azure. Przeprowadzona analiza wyników umożliwia również dobranie odpowiedniej do określonych zastosowań konfiguracji serwerów w chmurze.

EN

The paper presents the results of performance tests for solving the traveling salesman problem with the use of a parallel version of a genetic algorithm on a local multi-core server, and in the Windows Azure cloud. The analysis of the obtained results allows the selection of a suitable configuration of the servers in the cloud.

2

Optimization of a modular neural network for pattern recognition using parallel genetic algorithm

Cárdenas M., Melin P., Cruz L.

Journal of Automation Mobile Robotics and Intelligent Systems

|

2011

|

Vol. 5, No. 1

77-84

EN

In this paper, the implementation of a Parallel Genetic Algorithm (PGA) for the training stage, and the optimi zation of a monolithic and modular neural network, for pattern recognition are presented. The optimization con sists in obtaining the best architecture in layers, and neu rons per layer achieving the less training error in a shor ter time. The implementation was performed in a multicore architecture, using parallel programming techniques to exploit its resources. We present the results obtained in terms of performance by comparing results of the training stage for sequential and parallel implementations.

3

A distributed implementation of parallel genetic algorithm for slope stability evaluation

Srokosz P. E., Tran Ch.

Computer Assisted Mechanics and Engineering Sciences

|

2010

|

Vol. 17, no. 1

13-26

EN

Parallel processing, the method of considering many small tasks to solve one large problem, has emerged as a key enabling technology in modern computing. Parallel computers can be simply classified into shared memory systems and distributed memory systems. The shared memory computers have a global memory attached to a number of processors enabling several processors to work concurrently on different parts of the same computation. A different approach towards building large parallel computers is to connect several processors via a network. Each processor has its own local memory. The cost of building these computers increases with the number of processors. The distributed memory multiprocessor systems are scalable over a wider range than the shared memory computers. There are many intermediate computer architectures, each with its distinct programming model. Common between them is the notion of message passing. In all parallel processing, data must be exchanged between cooperating tasks. Several research groups have developed software packages such as Parallel Virtual Machine (PVM), theMessage Passing Interface (MPI), and others. In this paper, hardware implementation of parallel information processing is introduced by application of a multicellular computer idea, in which working cells were composed of general purpose one-chip microcomputers. The influence of the cellular computer's structure size on quality and efficiency of calculations was analyzed. The optimal structure consisted of 4x4 cells which guaranteed achieving satisfactory recurrence of results for an assumed set of working parameters. This paper presents an idea and the results of trial computations regarding the problem of slope stability evaluation by variational calculus assisted by genetic algorithm.

4

HeuristicLab Grid - a flexible and extensible environment for parallel heuristic optimization

Wagner S., Affenzeller M.

Systems Science

|

2004

|

Vol. 30, no 4

103-110

EN

Heuristic optimization techniques turned out to be very well suited for attacking various kinds of problems. However, when it comes to practical applications like scheduling problems, route planning, etc., also these algorithms still suffer from a very long running time mainly due to the rather large problem instances relevant in real world applications. Consequently, parallel optimization methods like parallel Genetic Algorithms are widely used to overcome this handicap. In this paper, the authors present a new environment for parallel heuristic optimization based upon the already proposed HeuristicLab. In contrast to other existing grid computing or parallel optimization projects, HeuristicLab Grid offers the possibility of rapid and easy use of existing optimization algorithms and problems in a parallel way without the need of complex installation and maintenance.

5

Genetyczny algorytm równoległy dla pewnego permutacyjnego problemu przepływowego

Bożejko W., Wodecki M.

Automatyka / Akademia Górniczo-Hutnicza im. Stanisława Staszica w Krakowie

|

2003

|

T. 7, z. 1-2

51-57

PL

W pracy rozpatrujemy permutacyjny problem przepływowy z minimalizacją sumy czasów zakończenia zadań, w literaturze oznaczany przez F‌m‌Csum. Należy on do klasy problemów silnie NP-trudnych. Przedstawiamy algorytm równoległy jego rozwiązywania oparty na metodzie algorytmu genetycznego, w którym wykorzystano ideę zrównoleglenia opartą na migracyjnym modelu wyspowym. Wykonano wiele obliczeń na reprezentatywnej grupie przykładów zamieszczonych w pracy Taillarda. Wyniki obliczeniowe porównano z najlepszymi znanymi w literaturze. Dla algorytmu równoległego uzyskano nie tylko przyspieszenie czasu obliczeń, ale również poprawę jakości i stabilności (dyspersji) rozwiązań.

EN

In this paper we consider the permutation flow-shop sequencing problem with the objective of minimizing the sum of task's flowtime, known as F‌m‌Csum in literature. We present parallel genetic algorithm based on the island model of migration. By computer simulations on Taillard benchmarks and the best known results from literature we obtained not only acceleration of the computation's time, but better quality and stability of the results too.