2006 HPCMP Users Group Conference (HPCMP-UGC'06)最新文献

{"title":"Building and Running Application Codes on the ARL Linux Networx Cluster (JVN)","authors":"G. Petit","doi":"10.1109/HPCMP-UGC.2006.8","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.8","url":null,"abstract":"Porting, building and running application codes in the high-performance Linux cluster environment are different than the traditional high-performance computing environment with which many users are familiar. There are multiple compiling environments to choose from, along with their associated MPICH environments and optimized scientific libraries. Determining which of these environments is best suited to a user's application is not always apparent, as is defining the user's environment to properly access the chosen compiler and libraries during the build process. After successfully building an application, creating LSF batch run scripts that will successfully and efficiently run these newly built applications to solve user-defined problems has its own unique requirements inherent to the Linux environment. The objective of this paper is to provide the information necessary for new users to the Linux cluster to determine the most appropriate compiling environment for their application, as well as the means to access the compiler and associated libraries to successfully build and, if necessary, debug the desired application. Usage of the LSF batch system to submit batch jobs that will successfully and efficiently run on Linux cluster compute nodes will also be addressed. A discussion of each compiler's characteristics will provide a basis for users to determine which compiling system is most appropriate for their application. An introduction to the use of software \"modules\" will provide a way for users to easily tailor their environment to use the desired compiling environment for code development/building. Finally, LSF batch script examples will be used to demonstrate correct script usage. By better understanding the ARL Linux Networx cluster environment, new users will be able to become effective and productive users of the system more quickly. By minimizing the time required by users to port their codes onto a new platform and begin production jobs, the overall efficacy of the cluster's capability as a research tool is enhanced thereby increasing its overall value to the program","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131740622","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}

{"title":"The Increasing Use of Visualization in Ship Hydrodynamics","authors":"J. Gorski, M. Ebert, S. Brewton, P. Chang, R. Miller, J. Slomski, W. Smith, W. Wilson","doi":"10.1109/HPCMP-UGC.2006.71","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.71","url":null,"abstract":"Flow field visualization is an important part of the study of fluid dynamics and ship hydrodynamics. The field of computational fluid dynamics has provided an unprecedented ability to explore the hydrodynamics of marine vehicles through visualization. Many examples of this exist in the literature for steady flow field situations. However, unsteady visualization provides both challenges and opportunities to extract meaningful physical insight and information from computational simulations. This paper discusses some of these issues along with approaches being pursued to obtain adequate flow field information using remote high performance computing resources as well as concurrent visualization using local resources. A number of examples of flow field computations being pursued are discussed including: cavity flow, ship roll motions, trailing edge vortex shedding, ballast water exchange and crashback","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133158369","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}

{"title":"Accuracy and Performance Testing of Three-Dimensional Unsaturated Flow Finite Element Groundwater Programs on the Cray XT3 Using Analytical Solutions","authors":"F. Tracy","doi":"10.1109/HPCMP-UGC.2006.2","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.2","url":null,"abstract":"Because Richards' equation for describing unsaturated groundwater flow is highly nonlinear, it is challenging to test the accuracy and efficiency of parallel three-dimensional finite element groundwater programs used to model such activities as remediation of military sites and computing flows and pressures in levees broken by Hurricane Katrina. To aid in this testing, the author has derived steady-state and transient analytical solutions for unsaturated flow. With these solutions, both accuracy of the numerical solutions and the performance of the parallel versions of the computer programs may be easily tested. The other significant feature of these analytical solutions is that the amount of nonlinearity of the equation may be easily adjusted by a parameter a. This paper first presents the equations for the analytical solutions of steady-state and transient flow into a rectangular block of initially dry soil. It then uses the program FEMWATER as a benchmark to compute the numerical solutions corresponding to these analytical solutions. Accuracy results for various sizes of the mesh and numbers of processing elements (PEs) on the Cray XT3 are computed for different values of a. The number of nonlinear iterations for the steady-state problem is also reported. The result of the accuracy assessment is that as a is increased, the errors also increase significantly with even 512 PEs not being enough compute power to adequately solve the problem. Overall parallel performance and performance of the message passing interface calls for updating the ghost nodes and are also tested with results given. The result of this study is that scaling of a 1,000,000-element problem was good up to 256 PEs (64 PEs is the typical sweet spot for FEMWATER), and the performance of the ghost-node updating routine met or exceeded expectations","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123883551","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}

{"title":"Information Management Support for Interactive High-Performance Computing Frameworks","authors":"G. Ramseyer, V. Ross, R. Linderman","doi":"10.1109/HPCMP-UGC.2006.39","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.39","url":null,"abstract":"The information management capabilities of the Joint Battlespace Infosphere (JBI) and the evolving 100X JBI were evaluated to support a network of remote computers. The network consisted of an interactive testbed of distributed high performance computers (HPCs) across the Defense Research and Engineering Network (DREN). This Distributed Interactive High Performance Computing Testbed (DIHT) was developed to enable researchers to operate in near real time in a diverse HPC environment. This interactive environment included access to six remote Linux clusters, and allowed the rapid execution of parallelized software in real time. At peak times the DIHT is producing hundreds to thousands of information objects per second","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116911531","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}

{"title":"Sourcery VSIPL++ HPEC Benchmark Performance","authors":"J. Bergmann, D. McCoy","doi":"10.1109/HPCMP-UGC.2006.65","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.65","url":null,"abstract":"Sourcery VSIPL++ is a high-performance, parallel, multi-platform implementation of the VSIPL++ standard, an open-architecture API for vector, signal, and image processing. Sourcery VSIPL++ allows users to develop signal-processing applications that are portable across multiple platforms and scalable across different machine sizes. This reduces program risk by avoiding lock-in to a particular vendor, platform, or technology. It allows programs to choose the best available commercial-off-the-shelf (COTS) technology and still take advantage of the technology refresh cycle. It lowers schedule pressure by enabling development to start on low-cost commodity systems before embedded systems are even available and by allowing development to initially focus on correctness of serial algorithms before optimizing for parallel performance","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"81 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126394071","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}

{"title":"Computational Science and Engineering Advances Understanding of Complex Unsteady Flows in High Performance Fans and Compressors","authors":"S. Gorrell, F. Tsung, Jixian Yao, R. Vickery","doi":"10.1109/HPCMP-UGC.2006.15","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.15","url":null,"abstract":"The ability of high fidelity simulations to more accurately model complex flow phenomena in high performance military fans and compressors is demonstrated. Challenge Project CII continues to investigate how blade row interactions and distortion transfer impact fan and compressor performance. Simulations accomplished through this Challenge Project have produced improved geometric, physical, and numerical fidelity results than previously possible, allowing the design of next generation engines to achieve unprecedented performance","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130179362","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}

{"title":"Modeling Breaking Ship Waves for Design and Analysis of Naval Vessels","authors":"G. Weymouth, K. Hendrickson, D. Yue, D. Dommermuth, P. Adams, R. Hand","doi":"10.1109/HPCMP-UGC.2006.46","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.46","url":null,"abstract":"Prediction of the performance and non-acoustical signature of surface ships which feature such effects as breaking waves, spray and air entrainment is still beyond the capabilities of standard numerical solution methods. The near-field flow about a surface ship is characterized by complex physical processes such as: (i) spray sheet and jet formation; (ii) strong free-surface turbulence interactions with (large-amplitude) breaking waves; (Hi) air entrainment and bubble generation; and (iv) post-breaking turbulence and dissipation. These physical phenomena still require resolutions that are not feasible in practical engineering flows, despite continuing advances in computational resources. A two-pronged approach is proposed to develop methods to accurately predict these complex physical systems. First, physics-based closure models for steep breaking waves in the presence of turbulence are developed with results from high-resolution direct numerical simulations of the Navier-Stokes equations. Second, cutting-edge parallel computing capabilities and newly developed solution techniques are utilized to simulate the free-surface flow around naval combatants moving at high speed. Direct numerical simulation is used to simulate an ensemble of breaking waves at moderate Reynolds numbers. Information derived from the breaking events in this study is being used as a first step in evaluating closure models for inclusion in existing LES and \"off-the-shelf\" RANS capabilities. Using NFA (numerical flow analysis), full scale simulations of a DDG model were performed. The simulations capture such features as wave breaking and air entrainment which are quantitatively compared to experimental results","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129506659","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}

{"title":"Performance Modeling of Emerging HPC Architectures","authors":"N. Bhatia, S. Alam, J. Vetter","doi":"10.1109/HPCMP-UGC.2006.58","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.58","url":null,"abstract":"Current state-of-art HPCMP performance modeling techniques primarily rely on combining a performance profile of an application on a well-known HPC architecture, and the machine characteristics of an emerging architecture to project an application's performance on the emerging architecture. Existing profiling and tracing tools on well-known architectures are typically used to collect the necessary performance data by executing applications and benchmarks on available systems. Since the performance enhancing features of novel processing devices may be significantly different from a conventional microprocessor system, current performance modeling schemes have limited applicability on systems like the Cray X1E vector supercomputer and parallel systems with accelerator devices like Cray XD1, which contains FPGAs. We employ an application modeling paradigm that allows a user to develop not only \"architecture aware\" but also \"application aware\"performance models. We extend the modeling assertions (MA) framework that permits a user to develop multi-resolution, parameterized symbolic models. We demonstrate the application of our modeling scheme by augmenting the MA models with performance enhancing attributes of the Cray X1E Multistrearning Processors (MSPs). Using the extended MA framework, we develop symbolic performance models of critical code blocks of an HPCMP 71-06 benchmark called HYCOM - an ocean modeling code. By representing the code characteristics of the critical code blocks in terms of both unique architectural attributes and key input parameters of the HYCOM application, we manage to reduce and sustain performance prediction error rates to less than 30%","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123136637","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}

{"title":"Direct Phase-Resolved Simulations of Large-Scale Nonlinear Ocean Wave-Fields","authors":"Guangyu Wu, Yuming Liu, D. Yue","doi":"10.1109/HPCMP-UGC.2006.19","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.19","url":null,"abstract":"We develop and apply a direct phase-resolved simulation capability for large-scale nonlinear ocean wavefield evolution. Unlike phase-averaged models, we directly solve the primitive Euler equation and preserve the phase of the wavefield during its nonlinear evolution. As a result, detailed descriptions of the free surface elevation and the kinematics of the wavefield are obtained. Using the direct simulation capability, we investigate the validity and limitations of existing approximate theories and models for the prediction of ocean surface wave statistics under different physical parameters. In particular, with 256 processors on the ERDC Cray XT3, we directly compute the nonlinear evolution of a short-crested (three-dimensional) wavefield in a domain of 28.7kmtimes28.7km with evolution time up to 30 minutes, and analyze the occurrence statistics and spatial distribution of rogue waves in the wavefield. The effects of spectral bandwidth and angular spreading on wave statistics are investigated","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"25 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134217859","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}

{"title":"Development of Plasma-based Flow Control Techniques for Scramjet Flow Paths","authors":"D. Gaitonde, M. Lindsey","doi":"10.1109/HPCMP-UGC.2006.18","DOIUrl":"https://doi.org/10.1109/HPCMP-UGC.2006.18","url":null,"abstract":"An overview is presented of a comprehensive challenge effort titled \"Tip-to-tail Turbulent Scramjet Flowpath Simulation with MHD Energy Bypass\". Emphasis is placed on fully three-dimensional (3-D) results obtained at flight conditions. The goal is to employ high-fidelity simulations to lift critical constraints in scramjet development through revolutionary new concepts, which do not readily lend themselves to ground or flight test investigations. The central feature of these techniques is the deployment of 3-D electromagnetic fields to address the harsh airbreathing hypersonic environment. The theoretical model employed resolves the multi-fluid, multi-physics phenomena through a careful blend of first-principles and phenomenological elements chosen to discover new enabling physics. The resulting large and mathematically complex set of equations requires both advanced numerical methods and the massive systems available through the DoD HPC. By coupling results to powerful 3-D visualization techniques, the project has discovered previously unknown effects relating the magnetic, electric and velocity vector fields, as well as location and extent of the ionized region. This has enabled the elucidation of a simple set of principles for effective plasma-based flow control. Key among these are the thermodynamic balance between ponderomotive force and heating through competition between work and Ohmic dissipation and the discriminating effect of energy direction when electrodes are employed. These general observations are employed to develop methods to mitigate heat transfer, separation, mixing and energy management","PeriodicalId":173959,"journal":{"name":"2006 HPCMP Users Group Conference (HPCMP-UGC'06)","volume":"7 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115437287","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}