Abstract. This paper outlines a draft definition for EC++. EC++ is a framework within which paral... more Abstract. This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set ...
This paper presents results of multi-stage numerical optimisation of medium sized portfolios to b... more This paper presents results of multi-stage numerical optimisation of medium sized portfolios to both bond and equity markets using High Performance Computing. Constraints applied are both linear and non-linear and the optimisation process is designed to handle multiple markets in multiple countries, producing a series of alternative scenarios dependent upon a number of risk-return requirements. Portfolios are allowed to contain instruments for which stochastic or historic data (or some combination) is available. Comparative performance for both real and artificial data sets is discussed and extrapolation to very large datasets will be presented. The comparative benefits of the deployment of large scale High Performance Computing in this class of problem will be made. Some discussion of the problem of portfolio stability in the context of this model is also included.
This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to ... more This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to both bond and equity markets using Hi h Performance Computing. Constraints applied are both linear and non-linear and the optimisation pro cess is designed to handle multiple markets in multiple countries, producing a series of alternati ve scenarios dependent upon a number of risk-return requirements. Portfolios are allowed to contain ins truments for which stochastic or historic data (or some combination) is available. Comparative perform ance for both real and artificial data sets is discussed and extrapolation to very large datasets will be presented. The comparative benefits of the deployment of large scale High Performance Computin g in this class of problem will be made. Some discussion of the problem of portfolio stability in the context of this model is also included.
This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to ... more This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to both bond and equity markets using Hi gh Performance Computing. Constraints applied are both linear and non-linear and the optimisation pro cess is designed to handle multiple markets in multiple countries, producing a series of alternati ve scenarios dependent upon a number of risk-return requirements. Portfolios
This paper presents results of multi-stage numerical optimisation of medium sized financial portf... more This paper presents results of multi-stage numerical optimisation of medium sized financial portfolios using High Performance Computing. Linear and non-linear constraints are applied and the optimisation process is designed to handle multiple markets in multiple countries, ...
Non-umklapp overlap integrals are calculated for GaP via a k.p band structure approach. The resul... more Non-umklapp overlap integrals are calculated for GaP via a k.p band structure approach. The results are presented in graphical form and are included in a calculation of the Auger recombination coefficient for the band-band hole-hole-electron process in GaP and 300K. The resulting value B2=1.65*10-33 cm6 s-1, which compares favourably with recent approximation schemes, is taken to indicate that the process is not the dominant non-radiative recombination process at this temperature.
At Queen Mary, University of London, we have over twenty years of experience in Parallel Computin... more At Queen Mary, University of London, we have over twenty years of experience in Parallel Computing Applications, mostly on "massively parallel systems", such as the Distributed Array Processors (DAPs). The applications in which we were involved included design of numerical subroutine libraries, Finite Element software, graphics tools, the physics of organic materials, medical imaging, computer vision and more recently, Financial modelling. Two of the projects related to the latter are described in this paper, namely Portfolio Optimisation and Financial Risk Assessment.
This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ sy... more This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set of parallel computing models and paradigms as possible, both standard models and user extensible models. This is done entirely within standard C++, i.e. without syntactic extensions to C++.
The theory of a new method of calculating optical matrix elements for deep levels in semiconducto... more The theory of a new method of calculating optical matrix elements for deep levels in semiconductors utilising a lattice of delta -function impurity potentials is presented. The method is applied to gallium arsenide and the results thus obtained for the photoionisation cross section are shown to be in encouraging agreement with experiment. The calculations demonstrate clearly the change in the
An analytic model for the dielectric matrix in a non-simple metal is presented using parabolic ba... more An analytic model for the dielectric matrix in a non-simple metal is presented using parabolic bands to approximate the known bandstructure. General results are obtained and applied to the case of the transition metal nickel by using the resulting matrix to calculate the electron energy loss spectrum. Agreement between the predicted and observed spectra is found to be reasonable in
ABSTRACT Presents results for the case of a 4*4 impurity superlattice model of deep levels in sem... more ABSTRACT Presents results for the case of a 4*4 impurity superlattice model of deep levels in semiconductors. In particular the authors consider the case of gallium arsenide and make comparison with deep level optical spectroscopy measurements for the EL2 level and show the degree of agreement to be good. They also consider the effects of phonon coupling and show them to be important, in particular in the case of photoionisation to the valence band.
This paper outlines a draft de nition for EC++. EC++ is a framework within which parallel C++ sys... more This paper outlines a draft de nition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set of parallel computing models and paradigms as possible, both standard models and user extensible models. This is done entirely within standard C++, i.e. without syntactic extensions to C++.
Abstract. This paper outlines a draft definition for EC++. EC++ is a framework within which paral... more Abstract. This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set ...
This paper presents results of multi-stage numerical optimisation of medium sized portfolios to b... more This paper presents results of multi-stage numerical optimisation of medium sized portfolios to both bond and equity markets using High Performance Computing. Constraints applied are both linear and non-linear and the optimisation process is designed to handle multiple markets in multiple countries, producing a series of alternative scenarios dependent upon a number of risk-return requirements. Portfolios are allowed to contain instruments for which stochastic or historic data (or some combination) is available. Comparative performance for both real and artificial data sets is discussed and extrapolation to very large datasets will be presented. The comparative benefits of the deployment of large scale High Performance Computing in this class of problem will be made. Some discussion of the problem of portfolio stability in the context of this model is also included.
This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to ... more This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to both bond and equity markets using Hi h Performance Computing. Constraints applied are both linear and non-linear and the optimisation pro cess is designed to handle multiple markets in multiple countries, producing a series of alternati ve scenarios dependent upon a number of risk-return requirements. Portfolios are allowed to contain ins truments for which stochastic or historic data (or some combination) is available. Comparative perform ance for both real and artificial data sets is discussed and extrapolation to very large datasets will be presented. The comparative benefits of the deployment of large scale High Performance Computin g in this class of problem will be made. Some discussion of the problem of portfolio stability in the context of this model is also included.
This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to ... more This paper presents results of multi-stage numeric al optimisation of medium sized portfolios to both bond and equity markets using Hi gh Performance Computing. Constraints applied are both linear and non-linear and the optimisation pro cess is designed to handle multiple markets in multiple countries, producing a series of alternati ve scenarios dependent upon a number of risk-return requirements. Portfolios
This paper presents results of multi-stage numerical optimisation of medium sized financial portf... more This paper presents results of multi-stage numerical optimisation of medium sized financial portfolios using High Performance Computing. Linear and non-linear constraints are applied and the optimisation process is designed to handle multiple markets in multiple countries, ...
Non-umklapp overlap integrals are calculated for GaP via a k.p band structure approach. The resul... more Non-umklapp overlap integrals are calculated for GaP via a k.p band structure approach. The results are presented in graphical form and are included in a calculation of the Auger recombination coefficient for the band-band hole-hole-electron process in GaP and 300K. The resulting value B2=1.65*10-33 cm6 s-1, which compares favourably with recent approximation schemes, is taken to indicate that the process is not the dominant non-radiative recombination process at this temperature.
At Queen Mary, University of London, we have over twenty years of experience in Parallel Computin... more At Queen Mary, University of London, we have over twenty years of experience in Parallel Computing Applications, mostly on "massively parallel systems", such as the Distributed Array Processors (DAPs). The applications in which we were involved included design of numerical subroutine libraries, Finite Element software, graphics tools, the physics of organic materials, medical imaging, computer vision and more recently, Financial modelling. Two of the projects related to the latter are described in this paper, namely Portfolio Optimisation and Financial Risk Assessment.
This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ sy... more This paper outlines a draft definition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set of parallel computing models and paradigms as possible, both standard models and user extensible models. This is done entirely within standard C++, i.e. without syntactic extensions to C++.
The theory of a new method of calculating optical matrix elements for deep levels in semiconducto... more The theory of a new method of calculating optical matrix elements for deep levels in semiconductors utilising a lattice of delta -function impurity potentials is presented. The method is applied to gallium arsenide and the results thus obtained for the photoionisation cross section are shown to be in encouraging agreement with experiment. The calculations demonstrate clearly the change in the
An analytic model for the dielectric matrix in a non-simple metal is presented using parabolic ba... more An analytic model for the dielectric matrix in a non-simple metal is presented using parabolic bands to approximate the known bandstructure. General results are obtained and applied to the case of the transition metal nickel by using the resulting matrix to calculate the electron energy loss spectrum. Agreement between the predicted and observed spectra is found to be reasonable in
ABSTRACT Presents results for the case of a 4*4 impurity superlattice model of deep levels in sem... more ABSTRACT Presents results for the case of a 4*4 impurity superlattice model of deep levels in semiconductors. In particular the authors consider the case of gallium arsenide and make comparison with deep level optical spectroscopy measurements for the EL2 level and show the degree of agreement to be good. They also consider the effects of phonon coupling and show them to be important, in particular in the case of photoionisation to the valence band.
This paper outlines a draft de nition for EC++. EC++ is a framework within which parallel C++ sys... more This paper outlines a draft de nition for EC++. EC++ is a framework within which parallel C++ systems can be standardised. The framework will add portability to parallel C++ systems and will run across a variety of hardware architectures, while encompassing as wide a set of parallel computing models and paradigms as possible, both standard models and user extensible models. This is done entirely within standard C++, i.e. without syntactic extensions to C++.
Uploads
Papers by P. Dzwig