this article describes a new image decomposition strategy to use the GPU memory hierarchy architecture more efficiently. An image is decomposed into tiles and the size of each tile is determined based on the size of t...
详细信息
Performance and efficiency became recently key requirements of computer architectures. Modern computers incorporate Graphics processing Units (GPUs) into running data mining algorithms, as well as other general purpos...
详细信息
ISBN:
(纸本)9783642396403
Performance and efficiency became recently key requirements of computer architectures. Modern computers incorporate Graphics processing Units (GPUs) into running data mining algorithms, as well as other general purpose computations. In this paper, different parallelization methods are analyzed and compared in order to understand their applicability. From multi-threading on shared memory to using NVIDIA's GPU accelerators for increasing performance and efficiency on parallel computing, this work discusses the parallelization of data mining algorithms considering performance and efficiency issues. the performance is compared on both many-core systems and GPU accelerators on a distance measure algorithm using a relatively big data set. We optimize the way we deal with GPUs in heterogeneous systems to make them more suitable for big data mining applications with heavy distance calculations. Moreover, we focus on achieving a higher utilization of GPU resources and a better reuse of data. Our implementation of the content-based similarity algorithm SQFD on the GPU outperforms by up to 50x CPU counterparts, and up to 15x CPU multi-threaded implementations.
Timeliness, accuracy and effectiveness of manufacturing information in manufacturing and business process management have become important factors of constraint to business growth. Single RFID (Radio Frequency Identif...
详细信息
the proceedings contain 23 papers. the topics discussed include: efficient parallelalgorithms for XML filtering with structural and value constraints;a simulation-based method for eliciting requirements of online CIB...
ISBN:
(纸本)9783642366079
the proceedings contain 23 papers. the topics discussed include: efficient parallelalgorithms for XML filtering with structural and value constraints;a simulation-based method for eliciting requirements of online CIB systems;reducing latency and network load using location-aware memcache architectures;modeling capabilities as attribute-featured entities;governance policies for verification and validation of service choreographies;real-text dictionary for topic-specific web searching;evaluating cross-platform development approaches for mobile applications;information gathering tasks on the web: attempting to identify the user search behavior;web-based exploration of photos with time and geospace;mixed-initiative management of online calendars;knowledge discovery: data mining by self-organizing maps;and ranking location-dependent keywords to extract geographical characteristics from microblogs.
Withthe development of microarray technology, it is possible now to study and measure the expression profiles of thousands of genes simultaneously which can lead to identify subgroup of specific disease or extract hi...
详细信息
this Paper proposes an effective SoC hardware architecture implementing a VDP for Full HD TVs. the proposed architecture makes real time video processing possible with supporting efficient bus architecture and flexibl...
详细信息
this book constitutes the refereed proceedings of the 7thinternationalconference on Language and Automata theory and Applications, LATA 2013, held in Bilbao, Spain in April 2013. the 45 revised full papers presented...
ISBN:
(纸本)9783642370656
this book constitutes the refereed proceedings of the 7thinternationalconference on Language and Automata theory and Applications, LATA 2013, held in Bilbao, Spain in April 2013. the 45 revised full papers presented together with 5 invited talks were carefully reviewed and selected from 97 initial submissions. the volume features contributions from both classical theory fields and application areas (bioinformatics, systems biology, language technology, artificial intelligence, etc.). Among the topics covered are algebraic language theory; algorithms for semi-structured data mining; algorithms on automata and words; automata and logic; automata for system analysis and program verification; automata, concurrency and Petri nets; automatic structures; cellular automata; combinatorics on words; computability; computational complexity; computational linguistics; data and image compression; decidability questions on words and languages; descriptional complexity; DNA and other models of bio-inspired computing; document engineering; foundations of finite state technology; foundations of XML; fuzzy and rough languages; grammars (Chomsky hierarchy, contextual, multidimensional, unification, categorial, etc.); grammars and automata architectures; grammatical inference and algorithmic learning; graphs and graph transformation; language varieties and semigroups; language-based cryptography; language-theoretic foundations of artificial intelligence and artificial life; parallel and regulated rewriting; parsing; pattern recognition; patterns and codes; power series; quantum, chemical and optical computing; semantics; string and combinatorial issues in computational biology and bioinformatics; string processingalgorithms; symbolic dynamics; symbolic neural networks; term rewriting; transducers; trees, tree languages and tree automata; weighted automata.
In almost all language processing applications, languages are parsed employing classical algorithms (such as the LR(1) parsers generated by Bison), which are sequential due to their left-to-right state-dependent natur...
详细信息
Fresnel Seismic Tomography which uses a huge amount of seismic data is an efficient methodology of researching three-dimensional structure of earth. However, in practical application, it confronts with two key challen...
详细信息
ISBN:
(纸本)9780819495877
Fresnel Seismic Tomography which uses a huge amount of seismic data is an efficient methodology of researching three-dimensional structure of earth. However, in practical application, it confronts with two key challenges of enormous data volume and huge computation. It is difficult to accomplish computation tasks under normal operating environment and computation strategies. In this paper, a Job-By-Application parallel computation strategy, which uses MPI (Message Passing Interface) and Pthread hybrid programming models based on the cluster, is designed to implement Fresnel seismic tomography, this method can solve the problem of allocating tasks dynamically, improve the load balancing and scalability of the system effectively;and we adopted the cached I/O strategy to accommodate the limited memory resources. Experimental results demonstrated that the program implemented on these strategies could completed the actual job within the idea time, the running of the program was stable, achieved load balancing, showed a good speedup and could adapt to the hardware environment of insufficient memory.
Background: Gene regulatory networks (GRN) inference is an important bioinformatics problem in which the gene interactions need to be deduced from gene expression data, such as microarray data. Feature selection metho...
详细信息
Background: Gene regulatory networks (GRN) inference is an important bioinformatics problem in which the gene interactions need to be deduced from gene expression data, such as microarray data. Feature selection methods can be applied to this problem. A feature selection technique is composed by two parts: a search algorithm and a criterion function. Among the search algorithms already proposed, there is the exhaustive search where the best feature subset is returned, although its computational complexity is unfeasible in almost all situations. the objective of this work is the development of a low cost parallel solution based on GPU architectures for exhaustive search with a viable cost-benefit. We use CUDA (TM), a general purpose parallel programming platform that allows the usage of NVIDIA (R) GPUs to solve complex problems in an efficient way. Results: We developed a parallel algorithm for GRN inference based on multiple GPU cards and obtained encouraging speedups (order of hundreds), when assuming that each target gene has two multivariate predictors. Also, experiments using single and multiple GPUs were performed, indicating that the speedup grows almost linearly withthe number of GPUs. Conclusion: In this work, we present a proof of principle, showing that it is possible to parallelize the exhaustive search algorithm in GPUs with encouraging results. Although our focus in this paper is on the GRN inference problem, the exhaustive search technique based on GPU developed here can be applied (with minor adaptations) to other combinatorial problems.
暂无评论