Solutions to Parallel and Distributed Computing Problems: Lessons from Biological SciencesAlbert Y. Zomaya, Fikret Ercal, Stephan Olariu Solving problems in parallel and distributed computing through the use of bio-inspired techniques. Recent years have seen a surge of interest in computational methods patterned after natural phenomena, with biologically inspired techniques such as fuzzy logic, neural networks, simulated annealing, genetic algorithms, or evolutionary computer models increasingly being harnessed for problem solving in parallel and distributed computing. Solutions to Parallel and Distributed Computing Problems presents a comprehensive review of the state of the art in the field, providing researchers and practitioners with critical information on the use of bio-inspired techniques for improving software and hardware design in high-performance computing. Through contributions from top leaders in the field, this important book brings together current research results, exploring some of the most intriguing and cutting-edge topics from the world of biocomputing, including: * Parallel and distributed computing of cellular automata and evolutionary algorithms * How the speedup of bio-inspired algorithms will help their applicability in a wide range of problems * Solving problems in parallel simulation through such techniques as simulated annealing algorithms and genetic algorithms * Techniques for solving scheduling and load-balancing problems in parallel and distributed computers * Applying neural networks for problem solving in wireless communication systems |
Contents
LargeScale Simulation | 1 |
Parallel Implementations of Evolutionary Algorithms | 50 |
Toward Hybrid Biologically Inspired Heuristics | 69 |
Copyright | |
14 other sections not shown
Common terms and phrases
allocation application architecture assigned average behavior CA-based scheduler cell cellular automata chromosome clusters completion corresponding crossover operator data item data transfers defined Distributed Computing evaluation event evolutionary algorithms example execution Figure fitness function fitness value flow GA-based approach gdio genetic algorithm genetic operators global H. J. Siegel HC Kernel Heterogeneous Computing hybrid IEEE IEEE Trans implementation initial population input ITC cost iteration lattice load balancing machine matrix MDDG meta-task metaheuristic method migration Min-min multiprocessor multiprocessor scheduling mutation neighborhood neural networks neurons nodes NP complete number of processors off-line on-line optimal solutions optimistic optimization problems Parallel and Distributed parallel computing parallel genetic parallel simulation partitioning performance program graph randomly represents rollbacks rules scheduling algorithm scheduling problem scheduling string scheme Section selection shown in Fig simulated annealing solve speedup static subtasks synchronization tabu search task graph techniques total number vector