| Ananth Grama, Vipin Kumar, and Panos Pardalos. Parallel processing of discrete optimization problems. Working Paper. |
....problems using different search strategies and find it effective in obtaining linear or near linear speedups. Keywords: load balancing, threads, parallelism. 1. INTRODUCTION Parallel searches have proven to be very beneficial for the Artificial Intelligence and Operations Research communities [1][2] 3] Large combinatorial problems, which otherwise are impractical for a single processor, can be solved by exploiting the parallelism of multiprocessor machines. However, not all parallel searches are successful. A recognized measure of success of a parallel search is its speedup, the ....
....x c j n 1 j j = 5) subject to , 1 x a = j n 1 j ij i = 1 , m, 6) 0,1 x j j = 1, n. 7) Constraint (6) insures that each row is at least covered by one column and (7) is the integrality constraint. For this problem, we used a parallel integer linear programming model [1]. The global strategy was a best first search and the expansion of nodes by a thread was achieved using a linear solving engine, CPLEX 5.0. The nodes were stored in a central open list. The scheme was similar to ones described in [1] 17] However when the list became too large (memory resources ....
[Article contains additional citation context not shown here]
A. Grama, V. Kumar, and P. Pardalos, Parallel Processing of Discrete Optimization Problems, Encyclpedia of Microcomputers, John Wiley & Sons, 1993.
....conjunction represented by the square dot, we want to search the graph without visiting the same node more than once. Figure 1(b) illustrates such an example. To accelerate the performance of graph search, parallelizing the search has been studied for various discrete optimization problems [3, 6]. We will exploit this approach for searching the optimal conjunction. To avoid the repetition of visiting the same node, conventional graph search algorithms maintain the list of visited nodes [3, 6] which however could be a severe bottleneck of parallel search. We instead propose a rule of ....
....graph search, parallelizing the search has been studied for various discrete optimization problems [3, 6] We will exploit this approach for searching the optimal conjunction. To avoid the repetition of visiting the same node, conventional graph search algorithms maintain the list of visited nodes [3, 6], which however could be a severe bottleneck of parallel search. We instead propose a rule of rewriting a conjunction to others. We first apply the rewriting rule to the initial conjunction to obtain child conjunctions, and then we repeat application of the rule to descendant conjunctions so that ....
G. Y. Ananth, V. Kumar, and P. Pardalos. Parallel processing of discrete optimization problems. 1993.
No context found.
G.Y. Ananth, V. Kumar and P. M. Pardalos, Parallel Processing of Discrete Optimization Problems, In Encyclopedia of Microcomputers Vol. 13 (1993), pp. 129--147, Marcel Dekker Inc., New York.
No context found.
Ananth Grama, Vipin Kumar, and Panos Pardalos. Parallel processing of discrete optimization problems. Working Paper.
No context found.
Grama, A.; Kumar, V.; Pardalos, P.: Parallel Processing of Discrete Optimization Problems, Encyclopedia of Microcomputers, Wiley, New York, NY, 1993.
No context found.
A. Grama, V. Kumar and P. Pardalos. Parallel Processing of Discrete Optimization Problems. Encyclopedia of Microcomputers, Marcel Dekker Inc., New York, (1992).
No context found.
A. Grama, V. Kumar, and P.Pardalos. Parallel Processing of Discrete Optimization Problems. Encyclopaedia of Microcomputers, Marcel Dekker Inc., New York, #1992#.
No context found.
A. Grama, V. Kumar, and P.Pardalos. Parallel Processing of Discrete Optimization Problems. Encyclopaedia of Microcomputers, Marcel Dekker Inc., New York, (1992).
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC