In this paper we present an approach to parallelization of the program for computation of axisymmetrical forging process. the parallel algorithm we have applied is based on non-overlapping domain decomposition method....
详细信息
In this paper we present an approach to parallelization of the program for computation of axisymmetrical forging process. the parallel algorithm we have applied is based on non-overlapping domain decomposition method. A mesh of elements is divided into layers assigned to different processes. the parallel program was written in C using PVM and it was implemented on Convex Exemplar SPP1000 and on networked workstations IBM RS/6000-320. We have investigated dependence of performance of the elaborated parallel program on number of process and on number of nodes in the mesh.
Searching for the Longest Common Subsequence (LCS for short) is one of the most fundamental tasks in bioinformatics. In this paper, we present a parallel implementation of the LCS computation for heterogeneous master-...
详细信息
High resolution lattice-Boltzmann simulations of turbulent channel flow on the Quadrics parallel machine are presented. the parallel performance is discussed together with some preliminary results concerning the vorti...
详细信息
High resolution lattice-Boltzmann simulations of turbulent channel flow on the Quadrics parallel machine are presented. the parallel performance is discussed together with some preliminary results concerning the vorticity structures which appear near the wall layer and their influence on the scaling laws.
In biology, there is a research field 'bioinformatics' in which computers are used as a method for problem solving. Bioinformatics includes a topic that is related to the analysis of genetic information. To an...
详细信息
ISBN:
(纸本)9781538629963
In biology, there is a research field 'bioinformatics' in which computers are used as a method for problem solving. Bioinformatics includes a topic that is related to the analysis of genetic information. To analyze genetic information, a homology search is used. the homology search detects similar parts of two base sequences. the Smith-Waterman algorithm is one of the most famous approaches for a homology search. the algorithm is sometimes executed in parallel by using a GPU. In this paper, we propose an acceleration method of a parallel homology search based on the Smith-Waterman algorithm by using a GPU. the proposed method first reduces the range of calculation of the Smith-Waterman algorithm by preprocess. Next, it applies the Smith-Waterman algorithm in parallel to the multiple sub matrices obtained after the preprocess to improve efficiency. As a result of an evaluation experiment, we have measured that our method gained at most 3.8 times in execution speed compared to a simple parallel execution of the Smith-Waterman algorithm by a GPU. the gain is caused by reduction of the calculation of the Smith-Waterman algorithm and by detecting sub sequences that can be executed asynchronously in parallel.
the traditional mobile application development difficult, costly, cumber some management and other issues, the paper based on computing technology, combined withthe Android platform presented in English applied langu...
详细信息
ISBN:
(纸本)9781479969296
the traditional mobile application development difficult, costly, cumber some management and other issues, the paper based on computing technology, combined withthe Android platform presented in English applied language and computer linguistics online translation system design, improved the traditional English applied language application development for computer linguistics unified based on CCS(Cloud computing System) management of virtual resources to provide good network application services. this paper discusses the CL research in the background of Internet, including the new applications, resources, challenges and methodologies.
this paper presents how Self-Organizing Maps (SOMs) can be trained efficiently using several, simultaneously executing threads on a shared memory Symmetric MultiProcessing (SMP) computer. the training method is a batc...
详细信息
ISBN:
(纸本)354043786X
this paper presents how Self-Organizing Maps (SOMs) can be trained efficiently using several, simultaneously executing threads on a shared memory Symmetric MultiProcessing (SMP) computer. the training method is a batch version of the Tree-Structured Self-Organizing Map. We note that SMP type of parallel training is very useful for large data sets obtained from nature, the process industry or large document collections, since we do not encounter similar model size limitations as with hardware SOM implementations.
We address synchronization issues of some block matrix multiplication algorithms in a distributed computing environment. We discuss performance behavior of a client/server implementation of these algorithms focusing o...
详细信息
ISBN:
(纸本)3540341412
We address synchronization issues of some block matrix multiplication algorithms in a distributed computing environment. We discuss performance behavior of a client/server implementation of these algorithms focusing on the most appropriate version which delivers the minimum synchronization overhead. Numerical experiments are carried out using the NetSolve distributed computing system.
*A graph is a structure that can express the relationship between objects. the emergence of GNN enables deep learning to be applied in the field of graphs. However, most GNNs are trained offline and cannot be directly...
详细信息
ISBN:
(纸本)9781450395502
*A graph is a structure that can express the relationship between objects. the emergence of GNN enables deep learning to be applied in the field of graphs. However, most GNNs are trained offline and cannot be directly used in real-time monitoring scenarios such as financial risk control. In addition, due to the large scale of graph data, a single machine often cannot meet actual needs, and there are bottlenecks such as throughput performance. therefore, we propose a distributed graph inference computing framework, which can be applied to Encoder-Decoder GNN models. We complete the adaptation of the model by disassembling the graph data and using the extension storage and dynamic invocation mechanism to solve the model invocation problem. For inference performance, we implement dynamic graph construction through incremental composition and decouple the inference process to apply to different scenarios, so that GNNs conforming to the Encoder-Decoder style can be applied to the framework. A large number of experiments show that this method has good timeliness while improving the throughput upper limit, and can maintain the model effect of multi-tasking.
Even though many companies and nations are aware of the need for implementing mobile cloud computing to their business practice, sometimes it is difficult to clarify the objective of cloud computing service. the gener...
详细信息
ISBN:
(纸本)9781467308762
Even though many companies and nations are aware of the need for implementing mobile cloud computing to their business practice, sometimes it is difficult to clarify the objective of cloud computing service. the general trend is to just implement ASP (Active Server Provider) instead. the most noteworthy distributed computing framework in the field of cloud computing is 'Hadoop,' an open source tool developed by 'Yahoo' for large scale data processing and analysis. this research proposes a mobile cloud computing model, which utilizes open source codes from distributed computing frameworks, like Hadoop, in an attempt to improve the efficiency of business processing. We will also be able to study how to process and analyze unstructured data in parallel to this model, so that in the future, providing customized information for individuals may be possible using unstructured data.
In this paper, we propose a channel estimation technique to combat the rapidly time-varying characteristics of multipath channel. the proposed method uses a normalized least mean square (NLMS) based novel adaptation s...
详细信息
ISBN:
(纸本)9781424477425
In this paper, we propose a channel estimation technique to combat the rapidly time-varying characteristics of multipath channel. the proposed method uses a normalized least mean square (NLMS) based novel adaptation scheme with amplitude-division technique. It supposes multiple linear transversal filters as estimators, which are arranged in a parallel fashion. the coefficient vectors for each estimator are formed withthe amplitude-division based classification technique according to the information of the channel coefficient values. the coefficient vector selected at each iteration is adapted withthe NLMS algorithm to handle the time variation effect of the rapidly time-varying channel. Computer simulation results demonstrate that the proposed estimator provide better tracking performance than the conventional NLMS estimator and amplitude-division parallel LMS (ADPLMS) estimator for a second order Markov communication channel in various fade rate conditions.
暂无评论