Página 1 dos resultados de 593 itens digitais encontrados em 0.007 segundos

Modelo computacional baseado em técnicas wavelets para relacionar imagens digitais obtidas em diferentes escalas e resoluções; Computational model based on wavelet techniques for linking digital images obtained at different scales and resolutions

Minatel, Edson Roberto
Fonte: Biblioteca Digitais de Teses e Dissertações da USP Publicador: Biblioteca Digitais de Teses e Dissertações da USP
Tipo: Tese de Doutorado Formato: application/pdf
Publicado em 03/10/2003 PT
Relevância na Pesquisa
36.08%
É apresentado o desenvolvimento de um modelo computacional que visa relacionar imagens digitais obtidas em diferentes escalas e resoluções com aplicação de Wavelets. Seu desenvolvimento encontra-se no contexto multidisciplinar e situa-se na intersecção das linhas de pesquisa de áreas da Física, da Matemática e da Computação. Desta forma, optou-se na implementação por uma abordagem computacional dos estudos, com aplicação em imagens digitais provenientes da reconstrução de dados de tomografia computadorizada de Raios-X. Resultados indicam que a implementação do modelo computacional desenvolvido tem sua funcionalidade comprovada, uma vez que os atributos vetoriais dos objetos considerados para análise (poros) foram mantidos estáveis em diferentes resoluções estudadas. O modelo foi implementado em linguagem de programação C++ com uso de orientação a objetos e organizado em classes. Adicionalmente, sua aplicação é viabilizada para diversas plataformas computacionais no que tange a sistemas operacionais e processadores. Do ponto de vista científico, o sistema resultante, além de ser uma ferramenta importante no estudo de meios porosos através de imagens de tomografia computadorizada por Raios-X, contribui com métodos inovadores que fazem uso de Wavelets e são aplicados na suavização de bordas por técnica sub-pixel...

Computational effort analysis and control in High Efficiency Video Coding

Silva, Mateus Grellert da
Fonte: Universidade Federal do Rio Grande do Sul Publicador: Universidade Federal do Rio Grande do Sul
Tipo: Dissertação Formato: application/pdf
ENG
Relevância na Pesquisa
36.01%
Codificadores HEVC impõem diversos desafios em aplicações embarcadas com restrições computacionais, especialmente quando há restrições de processamento em tempo real. Para tornar a codificação de vídeos HEVC factível nessas situações, é proposto neste trabalho um Sistema de Controle de Complexidade (SCC) que se adapta dinamicamente a capacidades computacionais varáveis. Considera-se que o codificador faz parte de um sistema maior, o qual informa suas restrições como disponibilidade da CPU e processamento alvo para o SCC. Para desenvolver um sistema eficiente, uma extensiva análise de complexidade dos principais parâmetros de codificação é realizada. Nessa análise, foi definida uma métrica livre de particularidades da plataforma de simulação, como hierarquia de memória e acesso concorrente à unidade de processamento. Essa métrica foi chamada de Complexidade Aritmética e pode ser facilmente adaptada para diversas plataformas. Os resultados mostram que o SCC proposto atinge ganhos médios de 40% em complexidade com penalidade mínima em eficiência de compressão e qualidade. As análises de adaptabilidade e controlabilidade mostraram que o SCC rapidamente se adapta a diferentes restrições, por exemplo...

ANN statistical image recognition method for computer vision in agricultural mobile robot navigation

Lulio, Luciano C.; Tronco, Mario L.; Porto, Arthur J. V.
Fonte: Universidade Estadual Paulista Publicador: Universidade Estadual Paulista
Tipo: Conferência ou Objeto de Conferência Formato: 1771-1776
ENG
Relevância na Pesquisa
46.19%
The main application area in this project, is to deploy image processing and segmentation techniques in computer vision through an omnidirectional vision system to agricultural mobile robots (AMR) used for trajectory navigation problems, as well as localization matters. Thereby, computational methods based on the JSEG algorithm were used to provide the classification and the characterization of such problems, together with Artificial Neural Networks (ANN) for image recognition. Hence, it was possible to run simulations and carry out analyses of the performance of JSEG image segmentation technique through Matlab/Octave computational platforms, along with the application of customized Back-propagation Multilayer Perceptron (MLP) algorithm and statistical methods as structured heuristics methods in a Simulink environment. Having the aforementioned procedures been done, it was practicable to classify and also characterize the HSV space color segments, not to mention allow the recognition of segmented images in which reasonably accurate results were obtained. © 2010 IEEE.

Scheduling (ir)regular applications on heterogeneous platforms

Mariano, Artur Miguel Matos
Fonte: Universidade do Minho Publicador: Universidade do Minho
Tipo: Dissertação de Mestrado
Publicado em 21/09/2012 ENG
Relevância na Pesquisa
36.38%
Dissertação de mestrado em Engenharia de Informática; Current computational platforms have become continuously more and more heterogeneous and parallel over the last years, as a consequence of incorporating accelerators whose architectures are parallel and different from the CPU. As a result, several frameworks were developed to aid to program these platforms mainly targeting better productivity ratios. In this context, GAMA framework is being developed by the research group involved in this work, targeting both regular and irregular algorithms to efficiently run in heterogeneous platforms. Scheduling is a key issue of GAMA-like frameworks. The state of the art solutions of scheduling on heterogeneous platforms are efficient for regular applications but lack adequate mechanisms for irregular ones. The scheduling of irregular applications is particularly complex due to the unpredictability and the differences on the execution time of their composing computational tasks. This dissertation work comprises the design and validation of a dynamic scheduler’s model and implementation, to simultaneously address regular and irregular algorithms. The devised scheduling mechanism is validated within the GAMA framework, when running relevant scientific algorithms...

How good are MatLab, Octave and Scilab for computational modelling?

Almeida,Eliana S. de; Medeiros,Antonio C; Frery,Alejandro C
Fonte: Sociedade Brasileira de Matemática Aplicada e Computacional Publicador: Sociedade Brasileira de Matemática Aplicada e Computacional
Tipo: Artigo de Revista Científica Formato: text/html
Publicado em 01/01/2012 EN
Relevância na Pesquisa
46.28%
In this article we test the accuracy of three platforms used in computational modelling: MatLab, Octave and Scilab, running on i386 architecture and three operating systems (Windows, Ubuntu and Mac OS). We submitted them to numerical tests using standard data sets and using the functions provided by each platform. A Monte Carlo study was conducted in some of the datasets in order to verify the stability of the results with respect to small departures from the original input. We propose a set of operations which include the computation of matrix determinants and eigenvalues, whose results are known. We also used data provided by NIST (National Institute of Standards and Technology), a protocol which includes the computation of basic univariate statistics (mean, standard deviation and first-lag correlation), linear regression and extremes of probability distributions. The assessment was made comparing the results computed by the platforms with certified values, that is, known results, computing the number of correct significant digits. Mathematical subject classification: Primary: 06B10; Secondary: 06D05.

Network-based drug discovery by integrating systems biology and computational technologies

Leung, Elaine L.; Cao, Zhi-Wei; Jiang, Zhi-Hong; Zhou, Hua; Liu, Liang
Fonte: Oxford University Press Publicador: Oxford University Press
Tipo: Artigo de Revista Científica
EN
Relevância na Pesquisa
26.28%
Network-based intervention has been a trend of curing systemic diseases, but it relies on regimen optimization and valid multi-target actions of the drugs. The complex multi-component nature of medicinal herbs may serve as valuable resources for network-based multi-target drug discovery due to its potential treatment effects by synergy. Recently, robustness of multiple systems biology platforms shows powerful to uncover molecular mechanisms and connections between the drugs and their targeting dynamic network. However, optimization methods of drug combination are insufficient, owning to lacking of tighter integration across multiple ‘-omics’ databases. The newly developed algorithm- or network-based computational models can tightly integrate ‘-omics’ databases and optimize combinational regimens of drug development, which encourage using medicinal herbs to develop into new wave of network-based multi-target drugs. However, challenges on further integration across the databases of medicinal herbs with multiple system biology platforms for multi-target drug optimization remain to the uncertain reliability of individual data sets, width and depth and degree of standardization of herbal medicine. Standardization of the methodology and terminology of multiple system biology and herbal database would facilitate the integration. Enhance public accessible databases and the number of research using system biology platform on herbal medicine would be helpful. Further integration across various ‘-omics’ platforms and computational tools would accelerate development of network-based drug discovery and network medicine.

Application of Genotyping-by-Sequencing on Semiconductor Sequencing Platforms: A Comparison of Genetic and Reference-Based Marker Ordering in Barley

Mascher, Martin; Wu, Shuangye; Amand, Paul St.; Stein, Nils; Poland, Jesse
Fonte: Public Library of Science Publicador: Public Library of Science
Tipo: Artigo de Revista Científica
Publicado em 03/10/2013 EN
Relevância na Pesquisa
26.27%
The rapid development of next-generation sequencing platforms has enabled the use of sequencing for routine genotyping across a range of genetics studies and breeding applications. Genotyping-by-sequencing (GBS), a low-cost, reduced representation sequencing method, is becoming a common approach for whole-genome marker profiling in many species. With quickly developing sequencing technologies, adapting current GBS methodologies to new platforms will leverage these advancements for future studies. To test new semiconductor sequencing platforms for GBS, we genotyped a barley recombinant inbred line (RIL) population. Based on a previous GBS approach, we designed bar code and adapter sets for the Ion Torrent platforms. Four sets of 24-plex libraries were constructed consisting of 94 RILs and the two parents and sequenced on two Ion platforms. In parallel, a 96-plex library of the same RILs was sequenced on the Illumina HiSeq 2000. We applied two different computational pipelines to analyze sequencing data; the reference-independent TASSEL pipeline and a reference-based pipeline using SAMtools. Sequence contigs positioned on the integrated physical and genetic map were used for read mapping and variant calling. We found high agreement in genotype calls between the different platforms and high concordance between genetic and reference-based marker order. There was...

Computational Analysis of Functional Single Nucleotide Polymorphisms Associated with the CYP11B2 Gene

Jia, Minyue; Yang, Boyun; Li, Zhongyi; Shen, Huiling; Song, Xiaoxiao; Gu, Wei
Fonte: Public Library of Science Publicador: Public Library of Science
Tipo: Artigo de Revista Científica
Publicado em 07/08/2014 EN
Relevância na Pesquisa
35.94%
Single nucleotide polymorphisms (SNPs) are the most common type of genetic variations in humans and play a major role in the genetics of human phenotype variation and the genetic basis of human complex diseases. Recently, there is considerable interest in understanding the possible role of the CYP11B2 gene with corticosterone methyl oxidase deficiency, primary aldosteronism, and cardio-cerebro-vascular diseases. Hence, the elucidation of the function and molecular dynamic behavior of CYP11B2 mutations is crucial in current genomics. In this study, we investigated the pathogenic effect of 51 nsSNPs and 26 UTR SNPs in the CYP11B2 gene through computational platforms. Using a combination of SIFT, PolyPhen, I-Mutant Suite, and ConSurf server, four nsSNPs (F487V, V129M, T498A, and V403E) were identified to potentially affect the structure, function, and activity of the CYP11B2 protein. Furthermore, molecular dynamics simulation and structure analyses also confirmed the impact of these nsSNPs on the stability and secondary properties of the CYP11B2 protein. Additionally, utilizing the UTRscan, MirSNP, PolymiRTS and miRNASNP, three SNPs in the 3′UTR region were predicted to exhibit a pattern change in the upstream open reading frames (uORF)...

A survey on platforms for big data analytics

Singh, Dilpreet; Reddy, Chandan K
Fonte: Springer International Publishing Publicador: Springer International Publishing
Tipo: Artigo de Revista Científica
EN
Relevância na Pesquisa
26.25%
The primary purpose of this paper is to provide an in-depth analysis of different platforms available for performing big data analytics. This paper surveys different hardware platforms available for big data analytics and assesses the advantages and drawbacks of each of these platforms based on various metrics such as scalability, data I/O rate, fault tolerance, real-time processing, data size supported and iterative task support. In addition to the hardware, a detailed description of the software frameworks used within each of these platforms is also discussed along with their strengths and drawbacks. Some of the critical characteristics described here can potentially aid the readers in making an informed decision about the right choice of platforms depending on their computational needs. Using a star ratings table, a rigorous qualitative comparison between different platforms is also discussed for each of the six characteristics that are critical for the algorithms of big data analytics. In order to provide more insights into the effectiveness of each of the platform in the context of big data analytics, specific implementation level details of the widely used k-means clustering algorithm on various platforms are also described in the form pseudocode.

Real-Time Scheduling of Embedded Applications on Multi-Core Platforms

Fan, Ming
Fonte: FIU Digital Commons Publicador: FIU Digital Commons
Tipo: Artigo de Revista Científica Formato: application/pdf
Relevância na Pesquisa
26.22%
For the past several decades, we have experienced the tremendous growth, in both scale and scope, of real-time embedded systems, thanks largely to the advances in IC technology. However, the traditional approach to get performance boost by increasing CPU frequency has been a way of past. Researchers from both industry and academia are turning their focus to multi-core architectures for continuous improvement of computing performance. In our research, we seek to develop efficient scheduling algorithms and analysis methods in the design of real-time embedded systems on multi-core platforms. Real-time systems are the ones with the response time as critical as the logical correctness of computational results. In addition, a variety of stringent constraints such as power/energy consumption, peak temperature and reliability are also imposed to these systems. Therefore, real-time scheduling plays a critical role in design of such computing systems at the system level. We started our research by addressing timing constraints for real-time applications on multi-core platforms, and developed both partitioned and semi-partitioned scheduling algorithms to schedule fixed priority, periodic, and hard real-time tasks on multi-core platforms. Then we extended our research by taking temperature constraints into consideration. We developed a closed-form solution to capture temperature dynamics for a given periodic voltage schedule on multi-core platforms...

Real-time scheduling of embedded applications on multi-core platforms

Fan, Ming
Fonte: FIU Digital Commons Publicador: FIU Digital Commons
Tipo: Artigo de Revista Científica
EN
Relevância na Pesquisa
26.22%
For the past several decades, we have experienced the tremendous growth, in both scale and scope, of real-time embedded systems, thanks largely to the advances in IC technology. However, the traditional approach to get performance boost by increasing CPU frequency has been a way of past. Researchers from both industry and academia are turning their focus to multi-core architectures for continuous improvement of computing performance. In our research, we seek to develop efficient scheduling algorithms and analysis methods in the design of real-time embedded systems on multi-core platforms. Real-time systems are the ones with the response time as critical as the logical correctness of computational results. In addition, a variety of stringent constraints such as power/energy consumption, peak temperature and reliability are also imposed to these systems. Therefore, real-time scheduling plays a critical role in design of such computing systems at the system level. ^ We started our research by addressing timing constraints for real-time applications on multi-core platforms, and developed both partitioned and semi-partitioned scheduling algorithms to schedule fixed priority, periodic, and hard real-time tasks on multi-core platforms. Then we extended our research by taking temperature constraints into consideration. We developed a closed-form solution to capture temperature dynamics for a given periodic voltage schedule on multi-core platforms...

Commodity cluster computing for computational chemistry

Hawick, K.; Grove, D.; Coddington, P.; Buntine, M.
Fonte: Internet Journal of Chemistry Publicador: Internet Journal of Chemistry
Tipo: Artigo de Revista Científica
Publicado em //2000 EN
Relevância na Pesquisa
36.1%
Access to high-performance computing power remains crucial for many computational chemistry problems. Unfortunately, traditional supercomputers or cluster computing solutions from commercial vendors remain very expensive, even for entry level configurations, and are therefore often beyond the reach of many small to medium-sized research groups and universities. Clusters of networked commodity computers provide an alternative computing platform that can offer substantially better price/performance than commercial supercomputers. We have constructed a networked PC cluster, or Beowulf, dedicated to computational chemistry problems using standard ab initio molecular orbital software packages such as Gaussian and GAMESS-US. This paper introduces the concept of Beowulf computing clusters and outlines the requirements for running the ab initio software packages used by computational chemists at the University of Adelaide. We describe the economic and performance trade-offs and design choices made in constructing the Beowulf system, including the choice of processors, networking, storage systems, operating system and job queuing software. Other issues such as throughput, scalability, software support, maintenance, and future trends are also discussed. We present some benchmark results for the Gaussian 98 and GAMESS-US programs...

Material instrucional apresentando conteúdos de métodos computacionais para o ensino de física; Instructional material presenting contents of computational methods for physics teaching

Betz, Michel Emile Marcel; Ribeiro-Teixeira, Rejane Maria
Fonte: Universidade Federal do Rio Grande do Sul Publicador: Universidade Federal do Rio Grande do Sul
Tipo: Artigo de Revista Científica Formato: application/pdf
POR
Relevância na Pesquisa
36.08%
É inegável a necessidade do uso de ferramentas computacionais no ensino de Física desde o nível básico. Por isso, é aconselhável que se promova o estudo envolvendo ferramentas e métodos computacionais já na formação de professores de Física, como nos cursos de Licenciatura, pois o fato de os professores não se sentirem confortáveis para incorporar metodologias envolvendo essas ferramentas em sua prática de ensino, muitas vezes, deve-se à falta ou ao desconhecimento de materiais instrucionais que lhes sirvam de motivação e orientação. Neste trabalho, descreve-se um material instrucional desenvolvido na forma de hipertexto, com o objetivo de contribuir para alterar esse quadro. São considerados alguns softwares úteis no ensino da Física, que oferecem recursos interessantes e estão disponíveis para os sistemas operacionais Linux e Windows. Os principais recursos de cada um dos softwares são apresentados a partir de uma situação-exemplo de Física, que serve como ilustração para demonstrar a sua utilização. O primeiro software apresentado é a planilha eletrônica e, a seguir, é estudado o software Modellus. Ao final da apresentação de cada um desses softwares é sugerido um exercício como uma tarefa a ser realizada. O material ainda aborda os softwares HotPotatoes e CmapTools. Grande parte do material instrucional aqui apresentado foi utilizada em disciplina de “Métodos Computacionais no Ensino de Física”...

Computational meta'omics for microbial community studies

Segata, Nicola; Boernigen, Daniela; Tickle, Timothy L; Morgan, Xochitl C; Garrett, Wendy S; Huttenhower, Curtis
Fonte: European Molecular Biology Organization Publicador: European Molecular Biology Organization
Tipo: Artigo de Revista Científica
EN_US
Relevância na Pesquisa
36.05%
Complex microbial communities are an integral part of the Earth's ecosystem and of our bodies in health and disease. In the last two decades, culture-independent approaches have provided new insights into their structure and function, with the exponentially decreasing cost of high-throughput sequencing resulting in broadly available tools for microbial surveys. However, the field remains far from reaching a technological plateau, as both computational techniques and nucleotide sequencing platforms for microbial genomic and transcriptional content continue to improve. Current microbiome analyses are thus starting to adopt multiple and complementary meta'omic approaches, leading to unprecedented opportunities to comprehensively and accurately characterize microbial communities and their interactions with their environments and hosts. This diversity of available assays, analysis methods, and public data is in turn beginning to enable microbiome-based predictive and modeling tools. We thus review here the technological and computational meta'omics approaches that are already available, those that are under active development, their success in biological discovery, and several outstanding challenges.

Experiences with Automated Build and Test for Geodynamics Simulation Codes

Heien, Eric M.; Miller, Todd L.; Gietzel, Becky; Kellogg, Louise H.
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Publicado em 04/09/2013
Relevância na Pesquisa
26.23%
The Computational Infrastructure for Geodynamics (CIG) is an NSF funded project that develops, supports, and disseminates community-accessible software for the geodynamics research community. CIG software supports a variety of computational geodynamic research from mantle and core dynamics, to crustal and earthquake dynamics, to magma migration and seismology. To support this type of project a backend computational infrastructure is necessary. Part of this backend infrastructure is an automated build and testing system to ensure codes and changes to them are compatible with multiple platforms and that the changes do not significantly affect the scientific results. In this paper we describe the build and test infrastructure for CIG based on the BaTLab system, how it is organized, and how it assists in operations. We demonstrate the use of this type of testing for a suite of geophysics codes, show why codes may compile on one platform but not on another, and demonstrate how minor changes may alter the computed results in unexpected ways that can influence the scientific interpretation. Finally, we examine result comparison between platforms and show how the compiler or operating system may affect results.

Sustainable Software Development for Next-Gen Sequencing (NGS) Bioinformatics on Emerging Platforms

Swenson, Shel; Simmhan, Yogesh; Prasanna, Viktor; Parashar, Manish; Riedy, Jason; Bader, David; Vuduc, Richard
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Relevância na Pesquisa
36.19%
DNA sequence analysis is fundamental to life science research. The rapid development of next generation sequencing (NGS) technologies, and the richness and diversity of applications it makes feasible, have created an enormous gulf between the potential of this technology and the development of computational methods to realize this potential. Bridging this gap holds possibilities for broad impacts toward multiple grand challenges and offers unprecedented opportunities for software innovation and research. We argue that NGS-enabled applications need a critical mass of sustainable software to benefit from emerging computing platforms' transformative potential. Accumulating the necessary critical mass will require leaders in computational biology, bioinformatics, computer science, and computer engineering work together to identify core opportunity areas, critical software infrastructure, and software sustainability challenges. Furthermore, due to the quickly changing nature of both bioinformatics software and accelerator technology, we conclude that creating sustainable accelerated bioinformatics software means constructing a sustainable bridge between the two fields. In particular, sustained collaboration between domain developers and technology experts is needed to develop the accelerated kernels...

A fully parallel, high precision, N-body code running on hybrid computing platforms

Capuzzo-Dolcetta, R.; Spera, M.; Punzo, D.
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Relevância na Pesquisa
35.94%
We present a new implementation of the numerical integration of the classical, gravitational, N-body problem based on a high order Hermite's integration scheme with block time steps, with a direct evaluation of the particle-particle forces. The main innovation of this code (called HiGPUs) is its full parallelization, exploiting both OpenMP and MPI in the use of the multicore Central Processing Units as well as either Compute Unified Device Architecture (CUDA) or OpenCL for the hosted Graphic Processing Units. We tested both performance and accuracy of the code using up to 256 GPUs in the supercomputer IBM iDataPlex DX360M3 Linux Infiniband Cluster provided by the italian supercomputing consortium CINECA, for values of N up to 8 millions. We were able to follow the evolution of a system of 8 million bodies for few crossing times, task previously unreached by direct summation codes. The code is freely available to the scientific community.; Comment: Paper submitted to Journal of Computational Physics consisting in 28 pages, 9 figures.The previous submitted version was lacking of the bibliography, for a Tex problem

A Domain Specific Approach to Heterogeneous Computing: From Availability to Accessibility

Inggs, Gordon; Thomas, David; Luk, Wayne
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Publicado em 21/08/2014
Relevância na Pesquisa
26.34%
We advocate a domain specific software development methodology for heterogeneous computing platforms such as Multicore CPUs, GPUs and FPGAs. We argue that three specific benefits are realised from adopting such an approach: portable, efficient implementations across heterogeneous platforms; domain specific metrics of quality that characterise platforms in a form software developers will understand; automatic, optimal partitioning across the available computing resources. These three benefits allow a development methodology for software developers where they describe their computational problems in a single, easy to understand form, and after a modeling procedure on the available resources, select how they would like to trade between various domain specific metrics. Our work on the Forward Financial Framework ($F^3$) demonstrates this methodology in practise. We are able to execute a range of computational finance option pricing tasks efficiently upon a wide range of CPU, GPU and FPGA computing platforms. We can also create accurate financial domain metric models of walltime latency and statistical confidence. Furthermore, we believe that we can support automatic, optimal partitioning using this execution and modelling capability.; Comment: Presented at First International Workshop on FPGAs for Software Programmers (FSP 2014) (arXiv:1408.4423)

Accelerating Cosmic Microwave Background map-making procedure through preconditioning

Szydlarski, Mikolaj; Grigori, Laura; Stompor, Radek
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Relevância na Pesquisa
35.94%
Estimation of the sky signal from sequences of time ordered data is one of the key steps in Cosmic Microwave Background (CMB) data analysis, commonly referred to as the map-making problem. Some of the most popular and general methods proposed for this problem involve solving generalised least squares (GLS) equations with non-diagonal noise weights given by a block-diagonal matrix with Toeplitz blocks. In this work we study new map-making solvers potentially suitable for applications to the largest anticipated data sets. They are based on iterative conjugate gradient (CG) approaches enhanced with novel, parallel, two-level preconditioners. We apply the proposed solvers to examples of simulated non-polarised and polarised CMB observations, and a set of idealised scanning strategies with sky coverage ranging from nearly a full sky down to small sky patches. We discuss in detail their implementation for massively parallel computational platforms and their performance for a broad range of parameters characterising the simulated data sets. We find that our best new solver can outperform carefully-optimised standard solvers used today by a factor of as much as 5 in terms of the convergence rate and a factor of up to $4$ in terms of the time to solution...

Quantum Monte Carlo for large chemical systems: Implementing efficient strategies for petascale platforms and beyond

Scemama, Anthony; Caffarel, Michel; Oseret, Emmanuel; Jalby, William
Fonte: Universidade Cornell Publicador: Universidade Cornell
Tipo: Artigo de Revista Científica
Relevância na Pesquisa
46.32%
Various strategies to implement efficiently QMC simulations for large chemical systems are presented. These include: i.) the introduction of an efficient algorithm to calculate the computationally expensive Slater matrices. This novel scheme is based on the use of the highly localized character of atomic Gaussian basis functions (not the molecular orbitals as usually done), ii.) the possibility of keeping the memory footprint minimal, iii.) the important enhancement of single-core performance when efficient optimization tools are employed, and iv.) the definition of a universal, dynamic, fault-tolerant, and load-balanced computational framework adapted to all kinds of computational platforms (massively parallel machines, clusters, or distributed grids). These strategies have been implemented in the QMC=Chem code developed at Toulouse and illustrated with numerical applications on small peptides of increasing sizes (158, 434, 1056 and 1731 electrons). Using 10k-80k computing cores of the Curie machine (GENCI-TGCC-CEA, France) QMC=Chem has been shown to be capable of running at the petascale level, thus demonstrating that for this machine a large part of the peak performance can be achieved. Implementation of large-scale QMC simulations for future exascale platforms with a comparable level of efficiency is expected to be feasible.