ATP binding cassette (ABC) transporters form a large class of proteins that use energy from ATP to power transport across biological membranes against the thermodynamic gradient. They consist of two water-soluble domains that are responsible for ATP hydrolysis and two transmembrane domains that form the transport pathway. In addition, they often interact with auxiliary domains or proteins that may regulate the transport process or deliver substrates. Recent crystal structures have revealed the molecular architecture of ABC transporters. We are using a combination of computational techniques to study the dynamics of ABC transporters, the possible mechanism by which the energy of ATP hydrolysis is used for transport, and the interactions between the different domains. We present an overview of our current understanding of the interactions between the domains of the vitamin Bu importer BtuCD, new simulation results from the BtuCD protein, and two models of the interactions of BtuCD with its substrate binding protein BtuF.
{"title":"Domain coupling in the ABC transporter system BtuCD/BtuF: molecular dynamics simulation, normal mode analysis and protein-protein docking","authors":"Christian Kandt, E. Oloo, D. Tieleman","doi":"10.1109/HPCS.2007.15","DOIUrl":"https://doi.org/10.1109/HPCS.2007.15","url":null,"abstract":"ATP binding cassette (ABC) transporters form a large class of proteins that use energy from ATP to power transport across biological membranes against the thermodynamic gradient. They consist of two water-soluble domains that are responsible for ATP hydrolysis and two transmembrane domains that form the transport pathway. In addition, they often interact with auxiliary domains or proteins that may regulate the transport process or deliver substrates. Recent crystal structures have revealed the molecular architecture of ABC transporters. We are using a combination of computational techniques to study the dynamics of ABC transporters, the possible mechanism by which the energy of ATP hydrolysis is used for transport, and the interactions between the different domains. We present an overview of our current understanding of the interactions between the domains of the vitamin Bu importer BtuCD, new simulation results from the BtuCD protein, and two models of the interactions of BtuCD with its substrate binding protein BtuF.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"132 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121220708","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}
In the context of computational grids, a metascheduler is the service responsible for scheduling jobs across many geographically distributed processor clusters. Typically, these software systems are complex and difficult to understand, resulting in initial designs which are ad hoc and suboptimal. This paper shows how a formal design methodology can be to used to better understand the relationships between metascheduler parameters, and thereby to achieve a well-designed metascheduler. Using a Plackett-Burman design, the methodology is demonstrated in the design of a knapsack grid metascheduler. The design is performed usiing efficiency and delay as the target variables.
{"title":"Improved Grid Metascheduler Design using the Plackett-Burman Methodology","authors":"D. Vanderster, N. Dimopoulos, R. Sobie","doi":"10.1109/HPCS.2007.20","DOIUrl":"https://doi.org/10.1109/HPCS.2007.20","url":null,"abstract":"In the context of computational grids, a metascheduler is the service responsible for scheduling jobs across many geographically distributed processor clusters. Typically, these software systems are complex and difficult to understand, resulting in initial designs which are ad hoc and suboptimal. This paper shows how a formal design methodology can be to used to better understand the relationships between metascheduler parameters, and thereby to achieve a well-designed metascheduler. Using a Plackett-Burman design, the methodology is demonstrated in the design of a knapsack grid metascheduler. The design is performed usiing efficiency and delay as the target variables.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"88 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124821653","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}
In this paper we demonstrate the use of the graphics processing unit (GPU) to accelerate evolutionary computation applications, in particular genetic programming approaches. We show that it is possible to get speed increases of several hundred times over a typical CPU implementation, catapulting GPU processing for these applications into the realm of HPC This increase in performance also extends to artificial developmental systems, where evolved programs are used to construct cellular systems. Feasibility of this approach to efficiently evaluate artificial developmental systems based on cellular automata is demonstrated.
{"title":"Fast Genetic Programming and Artificial Developmental Systems on GPUs","authors":"Simon Harding, W. Banzhaf","doi":"10.1109/HPCS.2007.17","DOIUrl":"https://doi.org/10.1109/HPCS.2007.17","url":null,"abstract":"In this paper we demonstrate the use of the graphics processing unit (GPU) to accelerate evolutionary computation applications, in particular genetic programming approaches. We show that it is possible to get speed increases of several hundred times over a typical CPU implementation, catapulting GPU processing for these applications into the realm of HPC This increase in performance also extends to artificial developmental systems, where evolved programs are used to construct cellular systems. Feasibility of this approach to efficiently evaluate artificial developmental systems based on cellular automata is demonstrated.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126769682","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}
The nonlinear simulation of interfacial instabilities in miscible displacements in porous media often requires sophisticated numerical algorithms as well as very fine spatial and temporal resolutions. In this study, Hartley transform based pseudo-spectral method is used to simulate time evolution of thermoviscous fingers in rectilinear geometry. The problem is formulated using continuity equation, Darcy's law, and volume-averaged forms of convection-diffusion equation for mass and energy balance. The numerical code is validated against established results for isothermal displacements. The effects of exponential dependence of viscosity on concentration and temperature, Lewis number, and porosity on the stability of the thermo-viscous flow are examined. It has been generally observed that at practical values of porosity and Lewis number, the thermal front always lags behind the fluid front and the instability is dominated by the viscosity variation due to concentration.
{"title":"Nonlinear Simulation of Thermo-viscous Fingering in Miscible Displacements in Porous Media","authors":"M. N. Islam, B. Maini, J. Azaiez","doi":"10.1109/HPCS.2007.21","DOIUrl":"https://doi.org/10.1109/HPCS.2007.21","url":null,"abstract":"The nonlinear simulation of interfacial instabilities in miscible displacements in porous media often requires sophisticated numerical algorithms as well as very fine spatial and temporal resolutions. In this study, Hartley transform based pseudo-spectral method is used to simulate time evolution of thermoviscous fingers in rectilinear geometry. The problem is formulated using continuity equation, Darcy's law, and volume-averaged forms of convection-diffusion equation for mass and energy balance. The numerical code is validated against established results for isothermal displacements. The effects of exponential dependence of viscosity on concentration and temperature, Lewis number, and porosity on the stability of the thermo-viscous flow are examined. It has been generally observed that at practical values of porosity and Lewis number, the thermal front always lags behind the fluid front and the instability is dominated by the viscosity variation due to concentration.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129522676","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}
Geoffroy R. Vallée, T. Naughton, Wesley Bland, S. Scott
To ensure quality, software development has to include testing mechanisms. OSCAR today supports several Linux distributions and several architectures. In such a context, the release cycle suffers of a important overhead created by the testing and stabilization phase. To address this issue, an approach is to implement a tool for automatic testing. This paper presents such a tool which is based on the OSCAR command line interface. This tool, based on system- level virtualization techniques, creates a virtual cluster to perform the test. This approach has the benefit of not corrupting the system of the physical machine and guarantee that the environment used for testing has not been corrupted before testing.
{"title":"Automatic Testing Tool for OSCAR Using System-level Virtualization","authors":"Geoffroy R. Vallée, T. Naughton, Wesley Bland, S. Scott","doi":"10.1109/HPCS.2007.9","DOIUrl":"https://doi.org/10.1109/HPCS.2007.9","url":null,"abstract":"To ensure quality, software development has to include testing mechanisms. OSCAR today supports several Linux distributions and several architectures. In such a context, the release cycle suffers of a important overhead created by the testing and stabilization phase. To address this issue, an approach is to implement a tool for automatic testing. This paper presents such a tool which is based on the OSCAR command line interface. This tool, based on system- level virtualization techniques, creates a virtual cluster to perform the test. This approach has the benefit of not corrupting the system of the physical machine and guarantee that the environment used for testing has not been corrupted before testing.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130131038","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}
Small scale turbulence in toroidal magnetic fusion experiments (Tokamaks) causes the rapid loss of heat from the plasma. This is of great importance since it limits the fusion performance of proposed reactor concepts. Electron temperature gradient driven (ETG) modes have been proposed as a source of anomalous electron thermal loses in tokamaks. It is widely acknowledged that the electrostatic potential in ETG turbulence can develop into radially elongated structures known as streamers. Understanding the conditions that permit streamers to produce experimentally significant transport is a topic of great interest. Analysis of the ETG mode at long wavelengths where both the ions and electrons are adiabatic (have a Boltzmann response) show that the ETG mode is inherently electromagnetic. Mixing length estimates of the thermal transport coefficient in this regime peak at collisionless skin-depth scales, providing a possible beta (the ratio of plasma pressure to magnetic pressure) dependence of the resulting transport. Preliminary nonlinear flux-tube simulations of the electromagnetic ETG mode produce large transport from the magnetic nonlinearity, while streamers in the electrostatic potential are still formed.
{"title":"Gyrokinetic simulation of micro-turbulence in magnetically confined plasmas","authors":"N. Joiner, A. Hirose, W. Dorland","doi":"10.1109/HPCS.2007.18","DOIUrl":"https://doi.org/10.1109/HPCS.2007.18","url":null,"abstract":"Small scale turbulence in toroidal magnetic fusion experiments (Tokamaks) causes the rapid loss of heat from the plasma. This is of great importance since it limits the fusion performance of proposed reactor concepts. Electron temperature gradient driven (ETG) modes have been proposed as a source of anomalous electron thermal loses in tokamaks. It is widely acknowledged that the electrostatic potential in ETG turbulence can develop into radially elongated structures known as streamers. Understanding the conditions that permit streamers to produce experimentally significant transport is a topic of great interest. Analysis of the ETG mode at long wavelengths where both the ions and electrons are adiabatic (have a Boltzmann response) show that the ETG mode is inherently electromagnetic. Mixing length estimates of the thermal transport coefficient in this regime peak at collisionless skin-depth scales, providing a possible beta (the ratio of plasma pressure to magnetic pressure) dependence of the resulting transport. Preliminary nonlinear flux-tube simulations of the electromagnetic ETG mode produce large transport from the magnetic nonlinearity, while streamers in the electrostatic potential are still formed.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"61 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121671819","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}
Grid computing is one of the main paradigms for resource-intensive scientific applications. It enables resource sharing and dynamic allocation of computational resources, thus increasing access to distributed data, promoting operational flexibility and collaboration, and allowing service providers to scale efficiently to meet variable demands. Large-scale grids are complex systems composed of many components from different domains. Quality of service (QoS) in such environments is an important issue due to the distributed nature of the services. In this paper we propose an allocation algorithm for matching service requests of Grid users with Grid services based on a QoS metric using either matchmaking or a well-tested genetic algorithm: NSGA-II. Experiments are performed and results are discussed for both approaches.
{"title":"Selection Algorithm for Grid Services based on a Quality of Service Metric","authors":"S. Ludwig, S. Reyhani","doi":"10.1109/HPCS.2007.28","DOIUrl":"https://doi.org/10.1109/HPCS.2007.28","url":null,"abstract":"Grid computing is one of the main paradigms for resource-intensive scientific applications. It enables resource sharing and dynamic allocation of computational resources, thus increasing access to distributed data, promoting operational flexibility and collaboration, and allowing service providers to scale efficiently to meet variable demands. Large-scale grids are complex systems composed of many components from different domains. Quality of service (QoS) in such environments is an important issue due to the distributed nature of the services. In this paper we propose an allocation algorithm for matching service requests of Grid users with Grid services based on a QoS metric using either matchmaking or a well-tested genetic algorithm: NSGA-II. Experiments are performed and results are discussed for both approaches.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"224 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132788028","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}
D. J. Ferreira, M. Dantas, A. R. Pinto, C. Montez, Martius Rodriguez
There has been a steady increase in the use of mobile computing, especially with appliances such as sensors. The main challenge in this scenario is to increase network sensor lifetime. Cluster computing integration with wireless sensor networks can represent an interesting answer for high-performance computing to monitor several environments. Computational resources available from cluster configurations are cost effective components to improve several classes of applications in many organizations. Scientific, industrial and commercial applications are more relying on cluster performance. OSCAR is a useful open software approach to manage cluster of workstations. In this article, we present a middleware design and implementation that integrates an OSCAR cluster configuration with a wireless sensor network environment.
{"title":"A Middleware for OSCAR and Wireless Sensor Network Environments","authors":"D. J. Ferreira, M. Dantas, A. R. Pinto, C. Montez, Martius Rodriguez","doi":"10.1109/HPCS.2007.5","DOIUrl":"https://doi.org/10.1109/HPCS.2007.5","url":null,"abstract":"There has been a steady increase in the use of mobile computing, especially with appliances such as sensors. The main challenge in this scenario is to increase network sensor lifetime. Cluster computing integration with wireless sensor networks can represent an interesting answer for high-performance computing to monitor several environments. Computational resources available from cluster configurations are cost effective components to improve several classes of applications in many organizations. Scientific, industrial and commercial applications are more relying on cluster performance. OSCAR is a useful open software approach to manage cluster of workstations. In this article, we present a middleware design and implementation that integrates an OSCAR cluster configuration with a wireless sensor network environment.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123363058","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}
C. Ollivier-Gooch, L. Diachin, M. Shephard, T. Tautges
Software for numerical solution of partial differential equations requires accessing, manipulating, and often modifying information about the geometry of the computational domain, the mesh used for the simulation, and discrete data stored on that mesh. Typically, applications programmers would prefer to avoid the difficulty and complexity of creating their own modules for tasks like interacting with multiple geometric modelers, mesh adaptation, and optimization algorithms, rightly preferring instead to focus on the problem physics and on studying the physical results that the code produces. Ideally, these modules would be provided by experts in CAD modeling, meshing, and optimization, and written so that they can use the application's data regardless of the data structures used by the application. This paper describes a language- and data-structure-independent interface supporting query and modification of mesh data conforming to a general abstract data model.
{"title":"A Language-Independent API for Unstructured Mesh Access and Manipulation","authors":"C. Ollivier-Gooch, L. Diachin, M. Shephard, T. Tautges","doi":"10.1109/HPCS.2007.4","DOIUrl":"https://doi.org/10.1109/HPCS.2007.4","url":null,"abstract":"Software for numerical solution of partial differential equations requires accessing, manipulating, and often modifying information about the geometry of the computational domain, the mesh used for the simulation, and discrete data stored on that mesh. Typically, applications programmers would prefer to avoid the difficulty and complexity of creating their own modules for tasks like interacting with multiple geometric modelers, mesh adaptation, and optimization algorithms, rightly preferring instead to focus on the problem physics and on studying the physical results that the code produces. Ideally, these modules would be provided by experts in CAD modeling, meshing, and optimization, and written so that they can use the application's data regardless of the data structures used by the application. This paper describes a language- and data-structure-independent interface supporting query and modification of mesh data conforming to a general abstract data model.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125934946","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}
Based on our previous work, an algorithm called AWS is proposed in this paper for a double-layer workflow scheduling approach for the grid. AWS aims to overcome difficulties brought about by the clustered resource distribution within the grid. It partitions a workflow graph according to features of available resource clusters and of the graph itself. It does not require detailed status information or control privilege on every grid resource for grid schedulers at the global Grid level. As a result, the dependence on grid information services is reduced and, at the same time, the higher priority of local resource management policies is respected. Experimental results show that AWS is adaptive to the grid circumstances and its performance approaches ideal global fine-granularity scheduling methods.
{"title":"An Adaptive Double-layer Workflow Scheduling Approach for Grid Computing","authors":"Fangpeng Dong, S. Akl","doi":"10.1109/HPCS.2007.6","DOIUrl":"https://doi.org/10.1109/HPCS.2007.6","url":null,"abstract":"Based on our previous work, an algorithm called AWS is proposed in this paper for a double-layer workflow scheduling approach for the grid. AWS aims to overcome difficulties brought about by the clustered resource distribution within the grid. It partitions a workflow graph according to features of available resource clusters and of the graph itself. It does not require detailed status information or control privilege on every grid resource for grid schedulers at the global Grid level. As a result, the dependence on grid information services is reduced and, at the same time, the higher priority of local resource management policies is respected. Experimental results show that AWS is adaptive to the grid circumstances and its performance approaches ideal global fine-granularity scheduling methods.","PeriodicalId":354520,"journal":{"name":"21st International Symposium on High Performance Computing Systems and Applications (HPCS'07)","volume":"77 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132181405","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：