Search results
21 – 30 of 804
To demonstrate the scalability of the genetic hybrid algorithm (GHA) in monitoring a local neural network algorithm for difficult non‐linear/chaotic time series problems.
Abstract
Purpose
To demonstrate the scalability of the genetic hybrid algorithm (GHA) in monitoring a local neural network algorithm for difficult non‐linear/chaotic time series problems.
Design/methodology/approach
GHA is a general‐purpose algorithm, spanning several areas of mathematical problem solving. If needed, GHA invokes an accelerator function at key stages of the solution process, providing it with the current population of solution vectors in the argument list of the function. The user has control over the computational stage (generation of a new population, crossover, mutation etc) and can modify the population of solution vectors, e.g. by invoking special purpose algorithms through the accelerator channel. If needed, the steps of GHA can be partly or completely superseded by the special purpose mathematical/artificial intelligence‐based algorithm. The system can be used as a package for classical mathematical programming with the genetic sub‐block deactivated. On the other hand, the algorithm can be turned into a machinery for stochastic analysis (e.g. for Monte Carlo simulation, time series modelling or neural networks), where the mathematical programming and genetic computing facilities are deactivated or appropropriately adjusted. Finally, pure evolutionary computation may be activated for studying genetic phenomena. GHA contains a flexible generic multi‐computer framework based on MPI, allowing implementations of a wide range of parallel models.
Findings
The results indicate that GHA is scalable, yet due to the inherent stochasticity of neural networks and the genetic algorithm, the scalability evidence put forth in this paper is only indicative. The scalability of GHA follows from maximal node intelligence allowing minimal internodal communication in problems with independent computational blocks.
Originality/value
The paper shows that GHA can be effectively run on both sequential and parallel platforms. The multicomputer layout is based on maximizing the intelligence of the nodes – all nodes are provided with the same program and the available computational support libraries – and minimizing internodal communication, hence GHA does not limit the size of the mesh in problems with independent computational tasks.
Details
Keywords
Robert G. Whirley, John O. Hallquist and Gerald L. Goudreau
Recent progress in element technology in large scale explicit finite element codes has opened the way for the solution of elastoplastic shell problems of unprecedented complexity…
Abstract
Recent progress in element technology in large scale explicit finite element codes has opened the way for the solution of elastoplastic shell problems of unprecedented complexity. This new capability has focused attention on the numerical issues involved in the implementation of elastoplastic material models for shells, particularly when vectorizable algorithms are required for supercomputer applications. This paper reviews four algorithms currently in the literature for plane stress and shell plasticity. First, each of the four methods is described in detail. Next, an accuracy analysis is presented for each algorithm for perfectly plastic, linear kinematic hardening, and linear isotropic hardening cases. Finally, a comparison is made of the relative computational efficiency of the four algorithms, and the importance of vectorization is illustrated.
Abstract
Details
Keywords
This paper aims to review current developments in the production of innovative systems in cybernetics and management science.
Abstract
Purpose
This paper aims to review current developments in the production of innovative systems in cybernetics and management science.
Design/methodology/approach
The approach takes the form of a general review and survey of selected research and development topics.
Findings
The paper illustrates the multi‐ and trans‐disciplinary nature of cybernetics, systems and management sciences.
Practical implications
The choice of reviews provides an awareness of the current trends in these areas of endeavour.
Originality/value
The reviews are selected from a global database and give a studied assessment of present‐day research and development initiatives.
Details
Keywords
To discuss a new parallel algorithmic platform (minlp_machine) for complex mixed‐integer non‐linear programming (MINLP) problems.
Abstract
Purpose
To discuss a new parallel algorithmic platform (minlp_machine) for complex mixed‐integer non‐linear programming (MINLP) problems.
Design/methodology/approach
The platform combines features from classical non‐linear optimization methodology with novel innovations in computational techniques. The system constructs discrete search zones around noninteger discrete‐valued variables at local solutions, which simplifies the local optimization problems and reduces the search process significantly. In complicated problems fast feasibility restoration may be achieved through concentrated Hessians. The system is programmed in strict ANSI C and can be run either stand alone or as a support library for other programs. File I/O is designed to recognize possible usage in both single and parallel processor environments. The system has been tested on Alpha, Sun and Linux mainframes and parallel IBM and Cray XT4 supercomputer environments. The constrained problem can, for example, be solved through a sequence of first order Taylor approximations of the non‐linear constraints and feasibility restoration utilizing Hessian information of the Lagrangian of the MINLP problem, or by invoking a nonlinear solver like SQP directly in the branch and bound tree. minlp_machine( ) has been tested as a support library to genetic hybrid algorithm (GHA). The GHA(minlp_machine) platform can be used to accelerate the performance of any linear or non‐linear node solver. The paper introduces a novel multicomputer partitioning of the discrete search space of genuine MINLP‐problems.
Findings
The system is successfully tested on a small sample of representative MINLP problems. The paper demonstrates that – through concurrent nonlinear branch and bound search – minlp_machine( ) outperforms some recent competing approaches with respect to the number of nodes in the branch and bound tree. Through parallel processing, the computational complexity of the local optimization problems is reduced considerably, an important aspect for practical applications.
Originality/value
This paper shows that binary‐valued MINLP‐problems will reduce to a vector of ordinary non‐linear programming on a suitably sized mesh. Correspondingly, INLP‐ and ILP‐problems will require no quasi‐Newton steps or simplex iterations on a compatible mesh.
Details
Keywords
NASA Langley Research Centre (LaRC) in Hampton, Virginia, U.S.A., has bought a Convex Preemptor 5000 Series realtime supercomputer. It will be used to support flight simulation…
Abstract
NASA Langley Research Centre (LaRC) in Hampton, Virginia, U.S.A., has bought a Convex Preemptor 5000 Series realtime supercomputer. It will be used to support flight simulation research programmes using the LaRC realtime simulation facility.
This paper aims to review current research and developments with particular reference to some global initiatives.
Abstract
Purpose
This paper aims to review current research and developments with particular reference to some global initiatives.
Design/methodology/approach
A general review and survey of selected research and development topics is given, and studies of some new projects and programmes discussed.
Findings
Illustrates the multi‐ and trans‐disciplinary natures of studies in cybernetics, systems and management science, with a view to further research and development activity.
Practical implications
The choice of reviews provides an awareness of current trends in these areas of endeavour.
Originality/value
The reviews are selected from a global database and give a studied‐assessment of current research and development initiatives.
Details
Keywords
AN economical solution to the difficult equations encountered in fluid dynamics has been found by the Aerodynamics Research Department of British Aerospace at Filton, Bristol with…
Abstract
AN economical solution to the difficult equations encountered in fluid dynamics has been found by the Aerodynamics Research Department of British Aerospace at Filton, Bristol with the purchase of a £100,000 Masscomp 32‐bit micro supercomputer.
Over the past eight years, the MELVYL catalog has become one of the largest public access catalogs in the world, and now plays a central role in providing access to the library…
Abstract
Over the past eight years, the MELVYL catalog has become one of the largest public access catalogs in the world, and now plays a central role in providing access to the library resources of the University of California. Currently, under heavy load, the MELVYL catalog supports many hundreds of simultaneous terminal connections, servicing over a quarter of a million queries a week and displaying more than two million records a week to its user community. This article discusses the history of the network that has supported the MELVYL catalog from the early days of its prototype to the present. It also describes both the current technical and policy issues that must be addressed as the network moves into the 1990s, and the roles that the network is coming to play in integrating local automation, the union catalog, access to resource databases, and other initiatives. Sidebars discuss the TCP/IP protocol suite, internet protocol gateways, and Telenet and related inter‐operability problems.
Sergio Botelho Junior and Bill O’Gorman
This paper aims to explore high performance computing (HPC) in the context of the South East region of Ireland, which hosts a publicly available HPC infrastructure, by identifying…
Abstract
Purpose
This paper aims to explore high performance computing (HPC) in the context of the South East region of Ireland, which hosts a publicly available HPC infrastructure, by identifying whether companies, especially small and medium enterprises (SMEs), are using, or are prepared to use, HPC to improve their business processes, expansion and sustainability. The result of the analysis provides region-specific guidelines that are meant to improve the HPC landscape in the region. The lessons learned from this research may apply to other similar, and developing, European regions.
Design/methodology/approach
This paper explores the use of HPC in the context of the South East region of Ireland and examines whether companies, especially SMEs, are benefiting from the use of publicly available HPC infrastructure in the region. This paper also provides a set of recommendations, of a policy nature, and required actions to increase HPC usage, based on the reality of the region. Therefore, the first step in the process was to understand the HPC landscape in the South East region of Ireland. Interviews were conducted with higher education institute (HEI) staff who were knowledgeable about the HPC infrastructure of their institutes and also about whether collaboration between the HEIs and businesses from the same region exists. The interview findings allowed the proposal of region-specific guidelines to improve the HPC landscape and collaboration in the region. The guidelines were analysed and refined in a focus group with key regional stakeholders from academia, industry and government, who have experience and expertise in high-technology transfer processes happening in the region.
Findings
The findings of the current study strongly suggest that HPC usage by SMEs in the South East region of Ireland is still incipient; and that HPC knowledge is currently inadequately transferred from the HEI hosting the HPC infrastructure to public and private sector organisations based in the region. The findings also demonstrate that there are no courses or training programmes available dedicated to HPC and that the level of collaboration between the HEI hosting the HPC infrastructure and industry in the region is minimal as regards HPC usage and projects. Therefore, there is a need to put specific targeted policies and actions, both from a regional government and HEI perspectives, in place to encourage SMEs to optimise their processes by using HPC.
Originality/value
This research is unique as it provides customised region-specific recommendations (RSR) and feasible actions to encourage industry, especially SMEs, to use HPC and collaborate around it. The literature review identified that there is a lack of studies that can inform policymakers to include HPC in their innovation agenda. Previous research studies specifically focussing on HPC policies are even more scarce. Most of the existing research pertaining to HPC focusses on the technical aspect of HPC; therefore, this research and paper bring a new dimension to existing HPC research. Even though this research was focussed on the South East of Ireland region, the model that generated the RSRs can be extrapolated and applied to other regions that need to develop their HPC landscape and the use of HPC among SMEs in their respective regions.
Details