The Experts below are selected from a list of 99 Experts worldwide ranked by ideXlab platform

Adrianus T. De Hoop - One of the best experts on this subject based on the ideXlab platform.

  • Electromagnetic Field Theory in ðN þ 1Þ-Space-Time: A Modern Time-Domain Tensor/Array Introduction A consistent tensor/Array Notation is used in this paper to present electromagnetic theory in ðN þ 1Þ-space-time; this leads to considerable simplifica
    2013
    Co-Authors: Adrianus T. De Hoop
    Abstract:

    In this paper, a modern time-domain introduction is presented for electromagnetic field theory in ðN þ 1Þ-space- time. It uses a consistent tensor/Array Notation that accom- modates the description of electromagnetic phenomena in N-dimensional space (plus time), a requirement that turns up in present-day theoretical cosmology, where a unified theory of electromagnetic and gravitational phenomena is aimed at. The standard vectorial approach, adequate for describing electro- magnetic phenomena in ð3 þ 1Þ-space-time, turns out to be not generalizable to ðN þ 1Þ-space-time for N > 3a nd the tensor/ Array approach that, in fact, has been introduced in Einstein's theory of relativity, proves, together with its accompanying Notation, to furnish the appropriate tools. Furthermore, such an approach turns out to lead to considerable simplifications, such as the complete superfluousness of standard vector cal- culus and the standard condition on the right-handedness of the reference frames employed. Since the field equations do no more than interrelate (in a particular manner) changes of the field quantities in time to their changes in space, only ele- mentary properties of (spatial and temporal) derivatives are needed to formulate the theory. The tensor/Array Notation furthermore furnishes indications about the structure of the field equations in any of the space-time discretization proce- dures for time-domain field computation. After discussing the field equations, the field/source compatibility relations and the constitutive relations, the field radiated by sources in an un- bounded, homogeneous, isotropic, lossless medium is de- termined. All components of the radiated field are shown to be expressible as elementary operations acting on the scalar Green's function of the scalar wave equation in ðN þ 1Þ-space- time. Time-convolution and time-correlation reciprocity re- lations conclude the general theory. Finally, two items on field computation are touched upon: the space-time-integrated field equations method of computation and the time-domain Cartesian coordinate stretching method for constructing per- fectly matched computational embeddings. The performance of these items is illustrated in a demonstrator showing the 1-D pulsed electric-current and magnetic-current sources excited wave propagation in a layered medium.

  • electromagnetic field theory in ðn þ 1þ space time a modern time domain tensor Array introduction a consistent tensor Array Notation is used in this paper to present electromagnetic theory in ðn þ 1þ space time this leads to considerable simplificati
    2013
    Co-Authors: Adrianus T. De Hoop
    Abstract:

    In this paper, a modern time-domain introduction is presented for electromagnetic field theory in ðN þ 1Þ-space- time. It uses a consistent tensor/Array Notation that accom- modates the description of electromagnetic phenomena in N-dimensional space (plus time), a requirement that turns up in present-day theoretical cosmology, where a unified theory of electromagnetic and gravitational phenomena is aimed at. The standard vectorial approach, adequate for describing electro- magnetic phenomena in ð3 þ 1Þ-space-time, turns out to be not generalizable to ðN þ 1Þ-space-time for N > 3a nd the tensor/ Array approach that, in fact, has been introduced in Einstein's theory of relativity, proves, together with its accompanying Notation, to furnish the appropriate tools. Furthermore, such an approach turns out to lead to considerable simplifications, such as the complete superfluousness of standard vector cal- culus and the standard condition on the right-handedness of the reference frames employed. Since the field equations do no more than interrelate (in a particular manner) changes of the field quantities in time to their changes in space, only ele- mentary properties of (spatial and temporal) derivatives are needed to formulate the theory. The tensor/Array Notation furthermore furnishes indications about the structure of the field equations in any of the space-time discretization proce- dures for time-domain field computation. After discussing the field equations, the field/source compatibility relations and the constitutive relations, the field radiated by sources in an un- bounded, homogeneous, isotropic, lossless medium is de- termined. All components of the radiated field are shown to be expressible as elementary operations acting on the scalar Green's function of the scalar wave equation in ðN þ 1Þ-space- time. Time-convolution and time-correlation reciprocity re- lations conclude the general theory. Finally, two items on field computation are touched upon: the space-time-integrated field equations method of computation and the time-domain Cartesian coordinate stretching method for constructing per- fectly matched computational embeddings. The performance of these items is illustrated in a demonstrator showing the 1-D pulsed electric-current and magnetic-current sources excited wave propagation in a layered medium.

  • Electromagnetic Field Theory in $(N+1)$ -Space-Time: A Modern Time-Domain Tensor/Array Introduction
    Proceedings of the IEEE, 2013
    Co-Authors: Adrianus T. De Hoop
    Abstract:

    In this paper, a modern time-domain introduction is presented for electromagnetic field theory in (N+1)-space-time. It uses a consistent tensor/Array Notation that accommodates the description of electromagnetic phenomena in N-dimensional space (plus time), a requirement that turns up in present-day theoretical cosmology, where a unified theory of electromagnetic and gravitational phenomena is aimed at. The standard vectorial approach, adequate for describing electromagnetic phenomena in (3+1)-space-time, turns out to be not generalizable to (N+1)-space-time for N >; 3 and the tensor/Array approach that, in fact, has been introduced in Einstein's theory of relativity, proves, together with its accompanying Notation, to furnish the appropriate tools. Furthermore, such an approach turns out to lead to considerable simplifications, such as the complete superfluousness of standard vector calculus and the standard condition on the right-handedness of the reference frames employed. Since the field equations do no more than interrelate (in a particular manner) changes of the field quantities in time to their changes in space, only elementary properties of (spatial and temporal) derivatives are needed to formulate the theory. The tensor/Array Notation furthermore furnishes indications about the structure of the field equations in any of the space-time discretization procedures for time-domain field computation. After discussing the field equations, the field/source compatibility relations and the constitutive relations, the field radiated by sources in an unbounded, homogeneous, isotropic, lossless medium is determined. All components of the radiated field are shown to be expressible as elementary operations acting on the scalar Green's function of the scalar wave equation in (N+1) -space-time. Time-convolution and time-correlation reciprocity relations conclude the general theory. Finally, two items on field computation are touched upon: the space-time-integrated field equations method of computation and the time-domain Cartesian coordinate stretching method for constructing perfectly matched computational embeddings. The performance of these items is illustrated in a demonstrator showing the 1-D pulsed electric-current and magnetic-current sources excited wave propagation in a layered medium.

Diego L Rodrigues - One of the best experts on this subject based on the ideXlab platform.

  • fortran legacy code performance optimization sequential and parallel processing with openmp
    Computer Science and Information Engineering, 2009
    Co-Authors: Fernando Gustavo Tinetti, Monica A Lopez, Pedro G Cajaraville, Diego L Rodrigues
    Abstract:

    Several optimization alternatives are presented for legacy FORTRAN 77 scientific programs, each one with a quantitative characterization in terms of performance gain. Initially, sequential optimization is focused on the analisys of Level 3 BLAS (Basic Linear Algebra Subroutines) utilization, since BLAS have several performance optimized implementations. Also, the Fortran 90/95 Array Notation is used as a code upgrade from Fortran 77 to Fortran 90/95and, also, to provide the compiler a better source code for performance optimization. Since the shared memory parallel computing model is widely available (multiple cores and/or processors), the analysis of possible parallel processing via OpenMP is presented, along with the performance gain in a specific case. Sequential optimization as well parallelization work is done on a real (production code) program: a weather climate model implemented about two decades ago and used for climate research.

  • CSIE (2) - Fortran Legacy Code Performance Optimization: Sequential and Parallel Processing with OpenMP
    2009 WRI World Congress on Computer Science and Information Engineering, 2009
    Co-Authors: Fernando Gustavo Tinetti, Monica A Lopez, Pedro G Cajaraville, Diego L Rodrigues
    Abstract:

    Several optimization alternatives are presented for legacy FORTRAN 77 scientific programs, each one with a quantitative characterization in terms of performance gain. Initially, sequential optimization is focused on the analisys of Level 3 BLAS (Basic Linear Algebra Subroutines) utilization, since BLAS have several performance optimized implementations. Also, the Fortran 90/95 Array Notation is used as a code upgrade from Fortran 77 to Fortran 90/95and, also, to provide the compiler a better source code for performance optimization. Since the shared memory parallel computing model is widely available (multiple cores and/or processors), the analysis of possible parallel processing via OpenMP is presented, along with the performance gain in a specific case. Sequential optimization as well parallelization work is done on a real (production code) program: a weather climate model implemented about two decades ago and used for climate research.

Chen Yifeng - One of the best experts on this subject based on the ideXlab platform.

  • Programming Heterogeneous Systems with Array Types
    2015 15th IEEE ACM International Symposium on Cluster Cloud and Grid Computing (CCGrid 2015), 2015
    Co-Authors: Cui Xiang, Li Xiaowen, Chen Yifeng
    Abstract:

    This paper describes the use of Array Notation called PArray in refinement of parallel programs concerning Array type that separates the physical data layout and logical structure of multi-dimensional data, and the control flow diversion of heterogeneous processor units. A case study on matrix multiplication demonstrates refinement of PArray programs: the code evolves from a simple single CPU-thread code to a multi-thread code on CPU/MIC and then a GPU code by modifying the Array types in only a few lines of code. A GPU-based SGEMM is implemented in PArray and achieves almost the same Gflops of CUBLAS 4.0 when testing on a single node of Tian-1A system. Because the code operates directly on the logical structure of Array, the same SGEMM code can work on different physical Array data layouts.EICPCI-S(ISTP)1067-107

Yifeng Chen - One of the best experts on this subject based on the ideXlab platform.

  • CCGRID - Programming Heterogeneous Systems with Array Types
    2015 15th IEEE ACM International Symposium on Cluster Cloud and Grid Computing, 2015
    Co-Authors: Xiang Cui, Yifeng Chen
    Abstract:

    This paper describes the use of Array Notation called PArray in refinement of parallel programs concerning Array type that separates the physical data layout and logical structure of multi-dimensional data, and the control flow diversion of heterogeneous processor units. A case study on matrix multiplication demonstrates refinement of PArray programs: the code evolves from a simple single CPU-thread code to a multi-thread code on CPU/MIC and then a GPU code by modifying the Array types in only a few lines of code. A GPU-based SGEMM is implemented in PArray and achieves almost the same Gflops of CUBLAS 4.0 when testing on a single node of Tian-1A system. Because the code operates directly on the logical structure of Array, the same SGEMM code can work on different physical Array data layouts.

Amodern Time-domain - One of the best experts on this subject based on the ideXlab platform.