A study on efficient visualization and real-time interactivity of large-scale scenes is discussed. Introducing parallel processing technology, we present a parallelizable strategy with the pipeline algorithm, realize ...
详细信息
ISBN:
(纸本)0780379292
A study on efficient visualization and real-time interactivity of large-scale scenes is discussed. Introducing parallel processing technology, we present a parallelizable strategy with the pipeline algorithm, realize this parallel algorithm based on shared-memory, and then apply this program to a test site, the Peking Olympic Games planning mixed scenes, including real-time rendering, dynamical texture loading, quick browsing and so on. The results show a running performance and real-time interactivity improvement of DEPS (Digital Earth Prototype System) when using this algorithm. The parallel program of this paper was developed and running on a Silicon Graphics multiprocessor, Onyx 3200, with four MIPS R12000 processors and InfiniteReality 3 graphic accelerator, under IRIX 6.5 operating system.
We present an optimal parallel selection algorithm on the EREW PRAM. This algorithm runs in O(log n) time with n/log n processors. This complexity matches the known lower bound for parallel selection on the EREW PRAM ...
详细信息
ISBN:
(纸本)9780898715385
We present an optimal parallel selection algorithm on the EREW PRAM. This algorithm runs in O(log n) time with n/log n processors. This complexity matches the known lower bound for parallel selection on the EREW PRAM model. We therefore close this problem which has been open for more than a decade.
Existing parallel association rule mining algorithms suffer from many problems when mining massive transactional datasets. One major problem is that most of the parallel algorithms for a shared nothing environment are...
详细信息
Existing parallel association rule mining algorithms suffer from many problems when mining massive transactional datasets. One major problem is that most of the parallel algorithms for a shared nothing environment are Apriori-based algorithms. Apriori-based algorithms are proven to be not scalable due to many reasons, mainly: (1) the repetitive I/O disk scans, (2) the huge computation and communication involved during the candidacy generation. This paper proposes a new disk-based parallel association rule mining algorithm called Inverted Matrix, which achieves its efficiency by applying three new ideas. First, transactional data is converted into a new database layout called Inverted Matrix that prevents multiple scanning of the database during the mining phase, in which finding globally frequent patterns could be achieved in less than a full scan with random access. This data structure is replicated among the parallel nodes. Second, for each frequent item assigned to a parallel node, a relatively small independent tree is built summarizing co-occurrences. Finally, a simple and non-recursive mining process reduces the memory requirements as minimum candidacy generation and counting is needed, and no communication between nodes is required to generate all globally frequent patterns.
作者：
L.FW. GoesC.A.P.S. MartinsElectrical Engineering
Computational and Digita Systems Laboratory Pontifical Catholic University of Minas Gerais Belo Horizonte Minas Gerais Brazil Computer Science Department
Elechical Engineering Computational and Digital System Laboratory Pontifical Catholic University of Minas Gerais Belo Horizonte Minas Gerais Brazil
In this paper, we devise and implement a parallel protocol reassembling algorithm in application layer for large scale network intruding. Conventional network intrusion detection systems (NIDS) scan the incoming IP pa...
详细信息
ISBN:
(纸本)0780377818
In this paper, we devise and implement a parallel protocol reassembling algorithm in application layer for large scale network intruding. Conventional network intrusion detection systems (NIDS) scan the incoming IP packets and judge the attack types by the sensitive information matching. In our algorithm the IP fragments and the TCP stream are reassembled into an entire datagram in the application layer in parallel which is searched for sensitive field. This increases performance even under hostile loads and enables efficient intrusion detection in high speed networks. A high-performance parallel protocol reassembling algorithm is presented and implemented with SMTP protocol.
Turbulent flow has a significantly higher drag than the corresponding laminar flow at the same flow conditions, and therefore incurs a significant penalty of increased fuel consumption due to the extra thrust required...
详细信息
ISBN:
(纸本)0780379241
Turbulent flow has a significantly higher drag than the corresponding laminar flow at the same flow conditions, and therefore incurs a significant penalty of increased fuel consumption due to the extra thrust required. One possible way of decreasing the drag is to apply surface suction to delay the transition from laminar to turbulent flow. In this paper an aerofoil with 3 non-overlapping panels covering up to 20% of chord for boundary layer transition control is considered. The problem is complicated by the fact that panels can change both their positions and lengths. The complexity of the optimization problem is such that it is not practical to perform the investigation using a single processor. A constrained global parallel algorithm based on a combination of deformed configuration methods and controlled random search method is developed. It is shown that for the problem considered, good solutions can be found efficiently.
This paper describes a software environment devised to support parallel and sequential discrete-event simulation. It provides assistance to the user in issues such as selection of the synchronization protocol to be us...
详细信息
This paper describes a software environment devised to support parallel and sequential discrete-event simulation. It provides assistance to the user in issues such as selection of the synchronization protocol to be used in the execution of the simulation of the model. The software framework has been built upon the bulk-synchronous model of parallel computing. The well-defined structure of this model allowed us to predict the running time cost of synchronization protocols in accordance with the particular work-load generated by the execution of the simulation model. We exploit this feature to automatically generate the simulation program.
We propose an efficient parallel algorithm with simple static and dynamic scheduling for generating combinations. It can use any number of processors (NPlesn-m+1) in order to generate the set of all combinations of C(...
详细信息
We propose an efficient parallel algorithm with simple static and dynamic scheduling for generating combinations. It can use any number of processors (NPlesn-m+1) in order to generate the set of all combinations of C(n,m). The main characteristic of this algorithm is to require no integer larger than n during the whole computation. The performance results show that even without a perfect load balance, this algorithm has very good performance, mainly when n is large. Besides, the dynamic algorithm presents a good performance on heterogeneous parallel platforms
In beam-beam macroparticle simulations for collider rings, the accurate determination of the incoherent spectrum and potentially unstable coherent modes requires (1) large numbers of collisions, and (2) accurate elect...
详细信息
In beam-beam macroparticle simulations for collider rings, the accurate determination of the incoherent spectrum and potentially unstable coherent modes requires (1) large numbers of collisions, and (2) accurate electric field solutions at each collision. On a single processor, a selfconsistent simulation typically uses a 2D model of the beam-beam interaction in order to achieve a reasonable computation time, however for the long (~0.3 m) bunches in the LHC we wish to include the third dimension in order to account for effects such as longitudinal motion, crossing angle, and the beam size and density variations. We describe here a parallel algorithm, developed with MPI on a small commodity Linux cluster, to extend our simulation code BeamX from 2D to 3D using longitudinal subdivision (slicing) of the bunches. Although this paper concentrates on the computing methods, some performance trials and example results will also be shown
Data cube construction is a commonly used operation in data warehouses. Because of the volume of data that is stored and analyzed in a data warehouse and the amount of computation involved in data cube construction, i...
详细信息
Data cube construction is a commonly used operation in data warehouses. Because of the volume of data that is stored and analyzed in a data warehouse and the amount of computation involved in data cube construction, it is natural to consider parallel machines for this operation. We address a number of algorithmic issues in parallel data cube construction. First, we present an aggregation tree for sequential (and parallel) data cube construction, which has minimally bounded memory requirements. An aggregation tree is parameterized by the ordering of dimensions. We present a parallel algorithm based upon the aggregation tree. We analyze the interprocessor communication volume and construct a closed form expression for it. We prove that the same ordering of the dimensions minimizes both the computational and communication requirements. We also describe a method for partitioning the initial array and prove that it minimizes the communication volume. Experimental results from implementation of our algorithms on a cluster of workstations validate our theoretical results
暂无评论