The ray tracing algorithm produces realistic high quality images, but it requires a long time for calculation on a single processor machine, which limits its practical use. With the development of distributed object a...
详细信息
The ray tracing algorithm produces realistic high quality images, but it requires a long time for calculation on a single processor machine, which limits its practical use. With the development of distributed object architectures, such as CORBA (Common Object Request Broker), the most promising way to improve ray traced picture productions seems to be parallelisation, which offers both increased CPU power and memory facilities. A natural way for parallelisation is to distribute pixels over the CORBA object system. However, since we want to deal with large scenes and making our algorithm functional over any parallel architecture system, scene objects have also to be distributed among processors, so a modified parallel algorithm is necessary. We propose a new method, which distributes the scene objects among processors (CORBA objects) according to their speed frequencies, which uses the object coherency property. Our approach of exploiting the bus CORBA gives very encouraging results.
Obtaining information of both action of multiple free-swimming robot fishes and environment quickly and accurately from image sequence is the foundation of making decision and control in a MRFS (multiple robot fishes ...
详细信息
ISBN:
(纸本)078037925X
Obtaining information of both action of multiple free-swimming robot fishes and environment quickly and accurately from image sequence is the foundation of making decision and control in a MRFS (multiple robot fishes cooperation system). The real-time vision subsystem of MRFS and its tracking strategy for multi-target are introduced in this paper. Combining features of the robot fish and location background, an adaptive segmentation algorithm based on hue histogram and saturation histogram is proposed, which can adapt the changing environmental conditions. At the same time, integrating with computer parallel processing technology, the overall tracking algorithms are optimized by means of MMX and SSE instructions. The designed visual subsystem has been implemented in MRFS, and the results have shown its effectiveness by successfully tracking multiple free-swimming fishes and obstacles.
This paper introduces a natural implementation for broadcasting data on Ethernet based clusters used for parallel computing. Initially, it will be shown that libraries for Message Passing between processes such as PVM...
详细信息
This paper introduces a natural implementation for broadcasting data on Ethernet based clusters used for parallel computing. Initially, it will be shown that libraries for Message Passing between processes such as PVM and implementations of MPI do not implement efficiently this operation or there is no reliability in terms of its performance. An implementation for broadcast messages is presented, taking into account the Ethernet based hardware layer found in most of the clusters used for parallel computing. The proposed implementation for broadcasting data is compared with the broadcast message available in PVM and LAM/MPI. Also, some comments are made about the proposed broadcast messages when the hardware layer is not Ethernet based. Finally, it will be shown by experimentation how the proposed broadcast message is used in the context of parallel linear algebra operations, specifically for parallel matrix multiplication.
Montgomery's trick is a well known technique for performing simultaneous inversions of several field elements. However, this technique is a strictly sequential algorithm. Here the authors introduced a parallel alg...
详细信息
Montgomery's trick is a well known technique for performing simultaneous inversions of several field elements. However, this technique is a strictly sequential algorithm. Here the authors introduced a parallel algorithm for performing simultaneous inversions of several finite field elements. The algorithm uses a binary tree and can perform inversions of 2 r elements using 3times2 r-1 multipliers in (r + 1) multiplication rounds and one inversion round. The authors also described how to modify the algorithm when less number of multipliers is available. This parallel algorithm is used to obtain a new parallel algorithm for elliptic curve scalar multiplication using a fixed base point. The scalar multiplication algorithm is resistant against simple power analysis (SPA) and can be implemented with different number of multipliers (2,4,8,...). Results show that implementation with 2 multipliers can lead to almost 40% speed-up over previously best known sequential SPA resistant algorithm
The knapsack problem is very important in cryptosystem and in number theory. We propose a new parallel algorithm for the knapsack problem where the method of divide and conquer is adopted. Basing on an EREW-SIMD machi...
详细信息
The knapsack problem is very important in cryptosystem and in number theory. We propose a new parallel algorithm for the knapsack problem where the method of divide and conquer is adopted. Basing on an EREW-SIMD machine with shared memory, the proposed algorithm utilizes O(2/sup n/4/)/sup 1-/spl epsiv// processors, 0/spl les//spl epsiv//spl les/1, and O(2/sup n/) memory to find a solution for the n-element knapsack problem in time O(2/sup n/4/ (2/sup n/4/)/sup /spl epsiv//). Thus the cost of the proposed parallel algorithm is O(2/sup n/), which is optimal, and an improved result over the past researches.
The following topics are dealt with: Grid and distributed computing; scheduling task systems; shared-memory multiprocessors; imaging and visualization; testing and debugging; performance analysis and real-time systems...
详细信息
The following topics are dealt with: Grid and distributed computing; scheduling task systems; shared-memory multiprocessors; imaging and visualization; testing and debugging; performance analysis and real-time systems; scheduling for heterogeneous resources; networking; peer-to-peer and mobile computing; compiler technology and run-time systems; load balancing; network routing; parallel programming models; parallel algorithms; scheduling and storage; parallel and distributed performance; software for high performance clusters; decentralized algorithms; multithreading and VLIW; parallel and distributed real-time systems; high-level parallel programming models and supportive environments; Java for parallel and distributed computing; nature inspired distributed computing; high performance computational biology; advances in parallel and distributed computational models; reconfigurable architectures; communication architecture for clusters; next generation systems; fault-tolerant parallel and distributed systems; wireless, mobile and ad hoc networks; parallel and distributed image processing, video processing, and multimedia; formal methods for parallel programming; Internet computing and e-commerce; parallel and distributed scientific and engineering computing with applications; massively parallel processing; performance modeling, evaluation, and optimization of parallel and distributed systems; and parallel and distributed systems: testing and debugging.
This paper describes an improved thinning algorithm for binary images. We improve thinning algorithm from the fundamental properties such as connectivity, one-pixel width, robust to noise and speed. In addition, in or...
详细信息
ISBN:
(纸本)0769519601
This paper describes an improved thinning algorithm for binary images. We improve thinning algorithm from the fundamental properties such as connectivity, one-pixel width, robust to noise and speed. In addition, in order to overcome information loss, we integrated the contour and skeleton of pattern and proposed the threshold way. Some fundamental requirements of thinning and the shape of pattern are preserved very well. Algorithm is very robust to noise and eliminate some spurious branch. Above all, it can overcome the loss of information in pattern. Experimental results show the performance of the proposed algorithm.
In this paper, we propose a parallel algorithm for mining maximal frequent itemsets from databases. A frequent itemset is maximal if none of its supersets is frequent. The new parallel algorithm is named parallel max-...
详细信息
In this paper, we propose a parallel algorithm for mining maximal frequent itemsets from databases. A frequent itemset is maximal if none of its supersets is frequent. The new parallel algorithm is named parallel max-miner (PMM), and it is a parallel version of the sequential max-miner algorithm by R.J. Bayardo (1998). Most of existing mining algorithms discover the frequent k-itemsets on the kth pass over the databases, and then generate the candidate (k + 1)-itemsets for the next pass. Compared to those level-wise algorithms, PMM looks ahead at each pass and prunes more candidate itemsets by checking the frequencies of their supersets. We implemented PMM on a cluster of workstations, and evaluated its performance for various cases. PMM demonstrated better performance than other sequential and parallel algorithms, and its performance is quite scalable, even when there are large maximal frequent itemsets (i.e. long patterns) in databases.
We propose two randomized protocols by which n (n not known) initially identical stations of a Packet Radio Network (PRN) are assigned ID numbers from 1 to n to distinguish them. They run regardless of the number of s...
详细信息
We propose two randomized protocols by which n (n not known) initially identical stations of a Packet Radio Network (PRN) are assigned ID numbers from 1 to n to distinguish them. They run regardless of the number of stations per channel. The first one is a naive protocol and is derived from recursive probabilistic divide-and-conquer techniques. It requires n/ln k broadcast rounds, where k is the number of communication channels. The second solution needs the well-known prefix sums algorithm and we show that in this scenario the described protocol terminates in 0(n/k) broadcast rounds on the average case whenever k less than or equal to n/ln n. These results are obtained by means of the average case analysis of algorithms, using probabilistic generating functions and formal methods. Surprisingly, our last protocol performs as well as the efficiency-oriented protocol of Hayashi et al. in [1,2], which depends on the number of stations per channel. And moreover, it can handle the case where k is an element of [n/3 Inn, n/ln n]. Copyright (C) 2003 John Wiley Sons, Ltd.
This paper constructs a class of parallel BDF algorithms for the systems of delay differential equations with explicit BDF methods as predictor and implicit BDF methods as corrector. In the paper the stability analysi...
详细信息
This paper constructs a class of parallel BDF algorithms for the systems of delay differential equations with explicit BDF methods as predictor and implicit BDF methods as corrector. In the paper the stability analysis is .discussed, and a sufficient condition under which the systems of delay differential equations are asymptotically stable is derived. It is easily inferred that the stability of the algorithm is controlled by the stability of the corresponding ordinary differential equations. Some results of numerical tests are also presented. From theoretical analysis and numerical tests the conclusion can be drawn that the algorithm is not only for linear problems, but also for nonlinear problems.
暂无评论