Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74149
R. Roble
The variable outputs of the Sun that influence the Earth's atmosphere include electromagnetic radiation, the solar wind, and energetic particles emitted by flares and other disturbances. Irregularities in the solar wind can also modulate galactic cosmic rays that impact the Earth. In addition, there are variations in the solar energy at the top of the atmosphere caused by long-term changes in the Earth's orbital elements. All of these variations are currently being imposed on an atmosphere whose composition and thermal structure is changing as a result of trace-gas increases from man's activities. There is need for a next-generation climate model that extends from the ground to the thermosphere that can be used to study the effects of solar variability on a changing atmosphere. Such a model will probably require an increase in present-day computational power by a factor of 10 to 100.<>
{"title":"Global change: the solar influence","authors":"R. Roble","doi":"10.1109/SUPERC.1988.74149","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74149","url":null,"abstract":"The variable outputs of the Sun that influence the Earth's atmosphere include electromagnetic radiation, the solar wind, and energetic particles emitted by flares and other disturbances. Irregularities in the solar wind can also modulate galactic cosmic rays that impact the Earth. In addition, there are variations in the solar energy at the top of the atmosphere caused by long-term changes in the Earth's orbital elements. All of these variations are currently being imposed on an atmosphere whose composition and thermal structure is changing as a result of trace-gas increases from man's activities. There is need for a next-generation climate model that extends from the ground to the thermosphere that can be used to study the effects of solar variability on a changing atmosphere. Such a model will probably require an increase in present-day computational power by a factor of 10 to 100.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"31 1-2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126082497","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74150
R. S. Chen, M.B. Fiering
The authors present a basic multivariate framework for assessing the interactions between human activities and environmental processes that illustrates the range of potential nonlinearities, feedbacks, and synergisms and some of the possible analytic complexities. They suggest that as the scientific community is increasingly called on not just to diagnose specific environmental problems but also to explore policy options and other management issues, much greater emphasis will be placed on quantitative assessment of the larger question of management of the global environment. It is in this relatively unexplored arena that the authors foresee many opportunities for multidisciplinary, multivariate assessments that would undoubtedly benefit from advances in supercomputing technologies.<>
{"title":"Global change-human interactions: a multivariate assessment framework","authors":"R. S. Chen, M.B. Fiering","doi":"10.1109/SUPERC.1988.74150","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74150","url":null,"abstract":"The authors present a basic multivariate framework for assessing the interactions between human activities and environmental processes that illustrates the range of potential nonlinearities, feedbacks, and synergisms and some of the possible analytic complexities. They suggest that as the scientific community is increasingly called on not just to diagnose specific environmental problems but also to explore policy options and other management issues, much greater emphasis will be placed on quantitative assessment of the larger question of management of the global environment. It is in this relatively unexplored arena that the authors foresee many opportunities for multidisciplinary, multivariate assessments that would undoubtedly benefit from advances in supercomputing technologies.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129850100","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74144
M. Ginsberg, J. Johnson
The cost of just one physical car crash experiment can range between $50000 and $750000 depending on whether the test is performed on a production model car or on a prototype vehicle. The high expense of impact testing and the availability of a variety of vector and parallel computers have motivated the development of sophisticated software to model such complex physical phenomena. An investigation has recently been initiated to improve the performance of such programs. The purpose of the research is to define and apply methodology to reduce execution time and improve transportability of physical impact simulation software for vector and parallel processors. The authors focus on the design and analysis of a set of relevant test problems for applicability in current and future computational environments with vector and/or parallel facilities. The attributes of each test problem are discussed, along with performance data using the DYNA3D nonlinear physical impact simulation program on a CRAY X-MP/24 supercomputer.<>
{"title":"Benchmarking the performance of physical impact simulation software on vector and parallel computers","authors":"M. Ginsberg, J. Johnson","doi":"10.1109/SUPERC.1988.74144","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74144","url":null,"abstract":"The cost of just one physical car crash experiment can range between $50000 and $750000 depending on whether the test is performed on a production model car or on a prototype vehicle. The high expense of impact testing and the availability of a variety of vector and parallel computers have motivated the development of sophisticated software to model such complex physical phenomena. An investigation has recently been initiated to improve the performance of such programs. The purpose of the research is to define and apply methodology to reduce execution time and improve transportability of physical impact simulation software for vector and parallel processors. The authors focus on the design and analysis of a set of relevant test problems for applicability in current and future computational environments with vector and/or parallel facilities. The attributes of each test problem are discussed, along with performance data using the DYNA3D nonlinear physical impact simulation program on a CRAY X-MP/24 supercomputer.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"3 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114185215","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74132
T. Holst
The field of computational fluid dynamics (CFD) using large-scale supercomputer applications is discussed. Formulational and computational requirements for the various governing equations, including the Euler and Navier-Stokes approaches, are examined for typical problems including the viscous flow field solution about a complete aerospace vehicle. Recent computed results and experimental comparisons are given to highlight the presentation. The future of CFD associated with three-dimensional applications is found to be rapidly expanding across a broad front, including internal and external flows and flows across the entire speed regime.<>
{"title":"Supercomputer applications in computational fluid dynamics","authors":"T. Holst","doi":"10.1109/SUPERC.1988.74132","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74132","url":null,"abstract":"The field of computational fluid dynamics (CFD) using large-scale supercomputer applications is discussed. Formulational and computational requirements for the various governing equations, including the Euler and Navier-Stokes approaches, are examined for typical problems including the viscous flow field solution about a complete aerospace vehicle. Recent computed results and experimental comparisons are given to highlight the presentation. The future of CFD associated with three-dimensional applications is found to be rapidly expanding across a broad front, including internal and external flows and flows across the entire speed regime.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116384189","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74148
A. Semtner, R. Chervin
A three-dimensional hydrodynamic model has been constructed for the purpose of simulating ocean circulation on modern supercomputing architectures. The model is designed to take advantage of faster clock speeds, increased numbers of processors, and enlarged memories of machines expected to be available over the next decade. The model allows realistic global simulations to be conducted in support of the Global Change Program. This involves resolving the energetic scales of motion, usually referred to as 'eddies'. Such global ocean modeling is essential for proper representation of the full range of oceanic and climatic phenomena. The first-ever global eddy-resolving ocean calculation is reported as a genuine breakthrough made possible by today's supercomputers.<>
{"title":"Breakthroughs in ocean and climate modeling made possible by supercomputers of today and tomorrow","authors":"A. Semtner, R. Chervin","doi":"10.1109/SUPERC.1988.74148","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74148","url":null,"abstract":"A three-dimensional hydrodynamic model has been constructed for the purpose of simulating ocean circulation on modern supercomputing architectures. The model is designed to take advantage of faster clock speeds, increased numbers of processors, and enlarged memories of machines expected to be available over the next decade. The model allows realistic global simulations to be conducted in support of the Global Change Program. This involves resolving the energetic scales of motion, usually referred to as 'eddies'. Such global ocean modeling is essential for proper representation of the full range of oceanic and climatic phenomena. The first-ever global eddy-resolving ocean calculation is reported as a genuine breakthrough made possible by today's supercomputers.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121571617","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74131
S. Tomboulian, C. Streett, M. Macaraeg
The authors address the issue of solving the time-dependent incompressible Navier-Stokes equations on the Connection Machine 2, for the problem of transition to turbulence on the steady flow in a channel. The spectral algorithm used serially requires O(N/sup 4/) operations; using the massive parallelism of the CM it becomes an O(N/sup 2/) problem, then solving the equations on an N*N*N grid. Preliminary timings of the code, written in Lisp, are included and compared with a corresponding code optimized for the Cray-2 for a 128*128*101 grid.<>
{"title":"Spectral solution of the incompressible Navier-Stokes equations on the Connection Machine 2","authors":"S. Tomboulian, C. Streett, M. Macaraeg","doi":"10.1109/SUPERC.1988.74131","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74131","url":null,"abstract":"The authors address the issue of solving the time-dependent incompressible Navier-Stokes equations on the Connection Machine 2, for the problem of transition to turbulence on the steady flow in a channel. The spectral algorithm used serially requires O(N/sup 4/) operations; using the massive parallelism of the CM it becomes an O(N/sup 2/) problem, then solving the equations on an N*N*N grid. Preliminary timings of the code, written in Lisp, are included and compared with a corresponding code optimized for the Cray-2 for a 128*128*101 grid.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"305 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133040883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74130
B. Boghosian, W. Taylor, D. Rothman
A cellular automaton (CA) recently developed by D.H. Rothman and J.M. Keller (1988) simulates the flow of two incompressible, immiscible, viscous fluids in two dimensions. This automaton has been simulated on the CM-2 Connection Machine using a sequence of logical operations and table lookups to determine the state of a CA site from its old state and those of its neighbors. The logical operations are performed in parallel by each of the Connection Machine processors, while the table lookups use the indirect addressing capabilities among groups of 32 processors. A description is given of CA fluids, including the issue of isotropy, the choice of a rule set, and the averaging procedure used to obtain hydrodynamical quantities. The CM-2 Connection Machine is then described, with emphasis on the indirect addressing capabilities of the machine. A complete description is also given of the Rothman-Keller model for two-phase flow. It is shown how the indirect addressing is used in the simulation algorithm, and how a symmetry in the dynamics is used to reduce the size of the lookup tables by a factor of six. A time sequence of results showing the separation of two immiscible phases from an initially homogenized state is presented.<>
{"title":"A cellular automata simulation of two-phase flow on the CM-2 Connection Machine computer","authors":"B. Boghosian, W. Taylor, D. Rothman","doi":"10.1109/SUPERC.1988.74130","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74130","url":null,"abstract":"A cellular automaton (CA) recently developed by D.H. Rothman and J.M. Keller (1988) simulates the flow of two incompressible, immiscible, viscous fluids in two dimensions. This automaton has been simulated on the CM-2 Connection Machine using a sequence of logical operations and table lookups to determine the state of a CA site from its old state and those of its neighbors. The logical operations are performed in parallel by each of the Connection Machine processors, while the table lookups use the indirect addressing capabilities among groups of 32 processors. A description is given of CA fluids, including the issue of isotropy, the choice of a rule set, and the averaging procedure used to obtain hydrodynamical quantities. The CM-2 Connection Machine is then described, with emphasis on the indirect addressing capabilities of the machine. A complete description is also given of the Rothman-Keller model for two-phase flow. It is shown how the indirect addressing is used in the simulation algorithm, and how a symmetry in the dynamics is used to reduce the size of the lookup tables by a factor of six. A time sequence of results showing the separation of two immiscible phases from an initially homogenized state is presented.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132173411","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74137
C. Westbrook, W. Pitz
The development of detailed chemical kinetic reaction mechanisms for analysis of autoignition and knocking of complex hydrocarbon fuels is described. The wide ranges of temperature and pressure encountered by end gases in automobile engine combustion chambers result in extreme demands on the reaction mechanisms intended to describe knocking conditions. The reactions and chemical species that are most important in each temperature and pressure regime are discussed, and the validation of these reaction mechanisms through comparison with idealized experimental results is described. The use of these mechanisms is illustrated through comparisons between computed results and experimental data obtained in actual knocking engines.<>
{"title":"Computer modeling of engine knock chemistry","authors":"C. Westbrook, W. Pitz","doi":"10.1109/SUPERC.1988.74137","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74137","url":null,"abstract":"The development of detailed chemical kinetic reaction mechanisms for analysis of autoignition and knocking of complex hydrocarbon fuels is described. The wide ranges of temperature and pressure encountered by end gases in automobile engine combustion chambers result in extreme demands on the reaction mechanisms intended to describe knocking conditions. The reactions and chemical species that are most important in each temperature and pressure regime are discussed, and the validation of these reaction mechanisms through comparison with idealized experimental results is described. The use of these mechanisms is illustrated through comparisons between computed results and experimental data obtained in actual knocking engines.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116150601","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74128
J. M. Swisshelm
A three-dimensional computational aerodynamics algorithm has been multitasked for efficient parallel execution on the Cray-2. It provides a means for examining the multitasking performance of a complete computational fluid dynamics (CFD) application code. An embedded zonal multigrid scheme is used to solve the Reynolds-averaged Navier-Stokes equations for an internal flow model problem. The explicit nature of each component of the method allows a spatial partitioning of the computational domain to achieve a well-balanced task load for multiple-instruction, multiple-data-stream (MIMD) computers with vector-processing capability. Experiments have been conducted with both two- and three-dimensional multitasked cases. The best speedup attained by an individual task group was 3.54 on four processors of the Cray-2, while the entire solver yielded a speedup of 2.67 on four processors for the three-dimensional case. The multiprocessing efficiency of various types of computational tasks is examined, performance on two Cray-2s with different memory access speeds is compared, and extrapolation to larger problems is discussed.<>
{"title":"Multitasking a three-dimensional Navier-Stokes algorithm on the Cray-2","authors":"J. M. Swisshelm","doi":"10.1109/SUPERC.1988.74128","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74128","url":null,"abstract":"A three-dimensional computational aerodynamics algorithm has been multitasked for efficient parallel execution on the Cray-2. It provides a means for examining the multitasking performance of a complete computational fluid dynamics (CFD) application code. An embedded zonal multigrid scheme is used to solve the Reynolds-averaged Navier-Stokes equations for an internal flow model problem. The explicit nature of each component of the method allows a spatial partitioning of the computational domain to achieve a well-balanced task load for multiple-instruction, multiple-data-stream (MIMD) computers with vector-processing capability. Experiments have been conducted with both two- and three-dimensional multitasked cases. The best speedup attained by an individual task group was 3.54 on four processors of the Cray-2, while the entire solver yielded a speedup of 2.67 on four processors for the three-dimensional case. The multiprocessing efficiency of various types of computational tasks is examined, performance on two Cray-2s with different memory access speeds is compared, and extrapolation to larger problems is discussed.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127525795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 1988-11-14DOI: 10.1109/SUPERC.1988.74141
G. A. Riccardi, P.H. Schow
Cluster analysis is an interdisciplinary study which involves the grouping of similar objects based on their measured attributes. The purpose of a cluster analysis is to investigate the structure and organization of the objects being studied. A description is given of the adaptation of the ISODATA clustering algorithm for vector supercomputer execution. On the CYBER 205, the algorithm runs 30 times faster than the original algorithm on the CYBER 205 using full automatic vectorization and 300 times faster than on a VAX 11/780. The major source of improvement over automatic vectorization is achieved by reorganizing the data structures used by the program. The modified algorithm yields increased performance on any vector computer.<>
{"title":"Adaptation of the ISODATA clustering algorithm for vector supercomputer execution","authors":"G. A. Riccardi, P.H. Schow","doi":"10.1109/SUPERC.1988.74141","DOIUrl":"https://doi.org/10.1109/SUPERC.1988.74141","url":null,"abstract":"Cluster analysis is an interdisciplinary study which involves the grouping of similar objects based on their measured attributes. The purpose of a cluster analysis is to investigate the structure and organization of the objects being studied. A description is given of the adaptation of the ISODATA clustering algorithm for vector supercomputer execution. On the CYBER 205, the algorithm runs 30 times faster than the original algorithm on the CYBER 205 using full automatic vectorization and 300 times faster than on a VAX 11/780. The major source of improvement over automatic vectorization is achieved by reorganizing the data structures used by the program. The modified algorithm yields increased performance on any vector computer.<<ETX>>","PeriodicalId":103561,"journal":{"name":"Proceedings Supercomputing Vol.II: Science and Applications","volume":"78 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1988-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133751939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：