This paper proposes different strategies of parallelizing a multi-criteria GRASP (greedy randomized adaptive search problem) algorithm. The parallel GRASP algorithm is applied to the multi-criteria minimum spanning tr...
详细信息
This paper proposes different strategies of parallelizing a multi-criteria GRASP (greedy randomized adaptive search problem) algorithm. The parallel GRASP algorithm is applied to the multi-criteria minimum spanning tree problem, which is NP-hard. In this problem, a vector of costs is defined for each edge of the graph and the goal is to find all the efficient or Pareto optimal spanning trees (solutions). Each process finds a subset of efficient solutions. These subsets are joined using different strategies to obtain the final set of efficient solutions. The multi-criteria GRASP algorithm with the different parallel strategies are tested on complete graphs with n = 20, 30 and 50 nodes and r = 2 and 3 criteria. The computational results show that the proposed parallel algorithms reduce the execution time and the results obtained by the sequential version were improved.
Ant colony optimization algorithms are intrinsically distributed algorithms where independent agents are in charge of building solutions collaboratively. Stigmergy or indirect communication is the way in which each ag...
详细信息
Ant colony optimization algorithms are intrinsically distributed algorithms where independent agents are in charge of building solutions collaboratively. Stigmergy or indirect communication is the way in which each agent learns from the experience of the whole colony. In this sense, explicit communication models of ACO can be defined directly, resulting in parallel algorithms of high numerical and real time efficiency. We do so in this work, and apply the resulting algorithms to the minimum tardy task problem (MTTP), a scheduling problem that has been faced with other meta-heuristics in the past. The aim of this article is to report experimental results on the behavior of three types of parallel ACO algorithms on large instances of the mentioned problems with the goal of improving existing solutions significantly.
The three dimensional discrete cosine transform (3D DCT) has been widely used in many applications such as video compression. On the other hand, the k-ary n-cube is one of the most popular interconnection networks use...
详细信息
The three dimensional discrete cosine transform (3D DCT) has been widely used in many applications such as video compression. On the other hand, the k-ary n-cube is one of the most popular interconnection networks used in many recent multicomputers. As direct calculation of 3D DCT is very time consuming, many researchers have been working on developing algorithms and special-purpose architectures for fast computation of 3D DCT. This paper proposes a parallel algorithm for efficient calculation of 3D DCT on the k-ary n-cube multicomputers. The time complexity of the proposed algorithm is of O(N) for an N times N times N input data cube while direct calculation of 3D DCT has a complexity of O(N 6 )
Since MODIS data have the feature of huge capacity and multi-spectrum, it needs too much time and frequent I/O operations to correct them by RS software. This paper proposes a parallel algorithm for MODIS Data based o...
详细信息
Since MODIS data have the feature of huge capacity and multi-spectrum, it needs too much time and frequent I/O operations to correct them by RS software. This paper proposes a parallel algorithm for MODIS Data based on triangulation network. The input images are divided into several pieces and each CPU processes a piece independently. By implementing the algorithm on a cluster system, the results show that, this parallel algorithm improves the efficiency of geometrical correction greatly.
Unmanned aerial vehicle (UAV) research is an increasingly important pillar of national security and military interest. A high fidelity discrete event simulation is prerequisite to any systems implementation. The synch...
详细信息
ISBN:
(纸本)9780780395190
Unmanned aerial vehicle (UAV) research is an increasingly important pillar of national security and military interest. A high fidelity discrete event simulation is prerequisite to any systems implementation. The synchronous parallel environment for emulation and discrete event simulation (SPEEDES) is a versatile and powerful tool that can be used for realization of this objective. A suite of five expert measures the efficiency a parallel UAV swarming SPEEDES application. Results indicate that the conservative time management produces more than twice the speed up as optimistic time management.
Mining association rules from large databases is an important problem in data mining. FP-growth is a powerful algorithm to mine frequent patterns and it is non-candidate generation algorithm using a special structure ...
详细信息
Mining association rules from large databases is an important problem in data mining. FP-growth is a powerful algorithm to mine frequent patterns and it is non-candidate generation algorithm using a special structure FP-tree. In order to enhance the efficiency of FP-grown algorithm, propose a novel parallel algorithm PFPTC to create sub FP-trees concurrently and a FP-tree merging algorithm called FP-merge, which can merge two FP-trees into one FP-tree. Also propose a new efficient algorithm QFP-growth, which can avoid bottleneck of FP-growth in generating a huge number of intermediate result.
In this paper, we develop parallel algorithms for pricing American-style Asian options employing binomial tree method. We describe the algorithm, explain the complexities, and study the performance. We have extended o...
详细信息
In this paper, we develop parallel algorithms for pricing American-style Asian options employing binomial tree method. We describe the algorithm, explain the complexities, and study the performance. We have extended our algorithm to handle Asian options with up to 10 underlying assets and shown that the multi-asset Asian options offer a better problem for parallel computation.
The area of path planning has received a great deal of attention recently. algorithms are required that can deliver optimal paths for robots to take over homogeneous or non-homogeneous terrain. Optimal paths may be th...
详细信息
The area of path planning has received a great deal of attention recently. algorithms are required that can deliver optimal paths for robots to take over homogeneous or non-homogeneous terrain. Optimal paths may be those that involve the shortest distance travelled, the least number of turns or the least number of ascents and descents. The often highly complex nature of terrains and the necessity for realtime solutions have lead to a requirement for the development of parallel algorithms. Such problems have been notoriously difficult to parallelise efficiently; indeed it has been said that an efficiency of 25-60% should be considered a success. In this paper we present a parallel algorithm for finding optimal paths over non-homogeneous terrain that demonstrates superlinear speed-up.
Protein-protein interactions are strictly correlated to the surface shape because beside a large number of structural amino acids composing the core there are few superficial amino acids that define the functionality....
详细信息
Protein-protein interactions are strictly correlated to the surface shape because beside a large number of structural amino acids composing the core there are few superficial amino acids that define the functionality. This study concerns the development of a tool that starting from the 3D atomic coordinates of a protein, as retrieved from the protein data bank (PDB), models the macromolecular surface in an implicit way using an approach that is more suitable for this kind of analysis than the parametric one. The marching cubes algorithm is used to process the volumetric description of the protein obtaining a precise representation of the corresponding surface. Because of the large amount of data to consider in studying whole protein families, this algorithm is implemented in parallel on a computer cluster to improve its performance. The parallel version of marching cubes is developed in ASSIST, a high level structured parallel programming system, obtaining a near optimal performance considering computational activities, and acceptable performance including I/O.
The clusters of SMP using fast networks, such as the Myricom's Myrinet, have emerged as important platforms for high performance computing. Although their peak advertised performance is very high, their real perfo...
详细信息
ISBN:
(纸本)354043786X
The clusters of SMP using fast networks, such as the Myricom's Myrinet, have emerged as important platforms for high performance computing. Although their peak advertised performance is very high, their real performance may be much lower than the peak advertised performance for many applications. To achieve high performance, we need to take advantages of both SMP and cluster architectures. Based on the HPM model for parallel computing, the performance of clusters of SMP systems is analyzed, and principles to optimize parallel algorithms (both from the paxallelism and locality point of view) are proposed. The influence of memory hierarchies on the performance is highly emphasized. Some practical examples on commercial clusters of SMPs systems Dawning D2000-2 and D3000 are also given.
暂无评论