Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “SGI”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2

Designing a High Performance OpenSHMEM Implementation using Universal Common Communication Substrate as a Communication Middleware

OpenSHMEM is an effort to standardize the well-known SHMEM parallel programming library. The project aims to produce an open-source and portable SHMEM API and is led by ORNL and UH. In this paper, we optimize the current OpenSHMEM reference implementa- tion, based on GASNet, to achieve higher performance characteristics. To achieve these desired performance characteristics, we have redesigned an important component of the OpenSHMEM implementation, the network layer, to leverage a low-level communication library designed for imple- menting parallel programming models called UCCS. In particular, UCCS provides an interface and semantics such as native atomic operations and remote memory operations to better support PGAS programming models, including OpenSHMEM. Through the use of microbenchmarks, we evaluate this new OpenSHMEM implementation on various network metrics, including the latency of point-to-point and collective operations. Furthermore, we compare the performance of our OpenSHMEM imple- mentation with the state-of-the-art SGI SHMEM. Our results show that the atomic operations of our OpenSHMEM implementation outperform SGI s SHMEM implementation by 3%. Its RMA operations outperform both SGI s SHMEM and the original OpenSHMEM reference implemen- tation by as much as 18% and 12% for gets, and as much as 83% and 53% for puts.

Welch, Donald A↗

Fueling Studies on the Lithium Tokamak Experiment

Lithium plasma facing components reduce the flux of "recycled" particles en- tering the plasma edge from the plasma facing components. This results in increased external fueling requirements and provides the opportunity to control the magnitude and distribution of the incoming particle flux. It has been pre- dicted that the plasma density profile will then be determined by the deposition profile of the external fueling, rather than dominated by the recycled particle flux. A series of experiments on the Lithium Tokamak Experiment demonstrate that lithium wall coatings facilitate control of the neutral and plasma particle in- ventories. With fresh lithium coatings and careful gas injection programming, over 90% of the injected particle inventory can be absorbed in the lithium wall during a discharge. Furthermore, dramatic changes in the fueling requirements and plasma parameters were observed when lithium coatings were applied. This is largely due to the elimination of water as an impurity on the plasma facing components. A Molecular Cluster Injector (MCI) was developed for the fueling of LTX plas- mas. The MCI uses a supersonic nozzle, cooled to liquid nitrogen temperatures, to create the conditions necessary for molecular cluster formation. It has been predicted that molecular clusters will penetrate deeper into plasmas than gas- phase molecules via a reduced ionization cross-section and by improving the collimation of the neutral jet. Using an electron beam diagnostic, the densities of the cryogenic MCI are measured to be an order of magnitude higher than in the room-temperature jets formed with the same valve pressure. This indi- cates increased collimation relative to what would be expected from ideal gas dynamics alone. A systematic study of the fueling efficiencies achieved with the LTX fueling systems is presented. The fueling efficiency of the Supersonic Gas Injector (SGI) is demonstrated to be strongly dependent on the distance between the nozzle and plasma edge. The fueling efficiency of the gas puffer was improved by the addition of a guide tube to keep the gas flow collimated and directed towards the plasma edge. Contrary to results reported on other devices, the MCI yields fueling efficiencies that are at best equal to the SGI. This suggests that any enhanced penetration from cluster formation is nullified by the reduced velocity of the cryogenic gas. Furthermore, at very high neutral fluxes, the fueling efficiency of the MCI dropped off dramatically, suggesting that the build-up of pressure in the scrape-off-layer is producing a self-shielding effect. All of these results are consistent with the idea that the best gas-based fueling source is one that brings highly directed neutrals to the plasma edge, without being scattered or ionized in the scrape-off-layer. The density profiles of LTX plasmas are measured during external fueling with the gas puffer, SGI, and cryogenic MCI. While the density profile amplitude increases with higher fueling rates, the shape of the profile remains constant. This suggests that particle transport is such that differences in the neutral deposition cannot be discerned. The density profiles are consistently hollow, suggesting a region of reduced particle transport in the core of LTX plasmas.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Size Distribution of Graphite Nodules in Hypereutectic Cast Irons of Varying Nodularity

An SGI was machined into 400 g cylindrical pieces and remelted in an electrical resistance furnace protected by Ar gas to produce materials ranging from SGI to CGI. The graphite morphology was controlled by varying the holding time at 1723 K (1450 °C) between 10 and 60 minutes. The discrete sectional size distribution of nodules by number density was measured on cross sections of the specimens and translated to volumetric distribution by volume fraction. Subpopulations of nodules were distinguished by fitting Gaussian distribution functions to the measured distribution. Primary and eutectic graphite, were found to account for most of the volume of nodular graphite in all cases. For holding times of 40 minutes and greater, corresponding to nodularity roughly below 40 pct, the primary subpopulation was very small and difficult to distinguish, leaving eutectic nodules as the dominant subpopulation. The mode and standard deviation of the two subpopulations were roughly independent of nodularity. Moreover, the nodular and vermicular graphite were segregated in the microstructure. In conclusion, the results suggest that the parallel development of the vermicular eutectic had small influence on the size distribution of eutectic graphite nodules.

36 MATERIALS SCIENCE↗

MINIMUM DUST ABUNDANCES FOR PLANETESIMAL FORMATION VIA SECULAR GRAVITATIONAL INSTABILITIES

We estimate minimum dust abundances required for secular gravitational instability (SGI) to operate at the midplane dust layer of protoplanetary disks. For SGI to be a viable process, the growth time of the instability T{sub grow} must be shorter than the radial drift time of the dust T{sub drift}. The growth time depends on the turbulent diffusion parameter {alpha} because the modes with short wavelengths are stabilized by turbulent diffusion. Assuming that turbulence is excited via the Kelvin-Helmholtz or streaming instabilities in the dust layer, and that its strength is controlled by the energy supply rate from dust accretion, we estimate the diffusion parameter and the growth time of the instability. The condition T{sub grow} < T{sub drift} requires that the dust abundance must be greater than a critical abundance Z{sub min}, which is a function of the Toomre parameter Q{sub g} and the aspect ratio h{sub g} /r of the gas disk. For a wide range of parameter space, the required dust abundance is less than 0.1. A slight increase in dust abundance opens a possible route for the dust to directly collapse to planetesimals.

79 ASTRONOMY AND ASTROPHYSICS↗

OpenSHMEM Implementation of IOR Benchmark

The benchmarking effort within the Computational Research & Development Programs at the Oak Ridge National Laboratory (ORNL) seeks to design and enable High Performance Computing (HPC) benchmarks and test suites. The work described in this report is a part of the effort focusing on the comparison and analysis of OpenSHMEM implementations using the Interleave Or Random (IOR) software for benchmarking parallel file system using POSIX, MPIIO, or HDF5 interfaces. We describe the effort to emulate the MPIIO parallel collective capabilities in the IOR benchmark using OpenSHMEM communication. One development effort was in emulating the MPI derived datatype used in the read/write operations and in setting the file view. Another effort was in implementing an internal cache in OpenSHMEM distributed shared memory to facilitate global collective I/O operations. Experiments comparing collective I/O in MPIIO implementations with the OpenSHMEM implementations were performed on the SGI Turing Cluster and the Cray XK7 Titan supercomputer at the Oak Ridge Leadership Computing Facility (OLCF). The preliminary results suggest that on the Cray XK7 Titan, the MPIIO implementations obtained higher write performance and the OpenSHMEM version obtained slightly higher read performance. On the SGI Turing Cluster, the MPIIO implementations obtained slightly higher performance over the OpenSHMEM implementations on large files.

97 MATHEMATICS AND COMPUTING↗

PLOT3D/AMES

Interactive graphics program helps scientists visualize computational fluid dynamics grids and solutions. PLOT3D/AMES offers multitude of functions and features and adapted for more types of computers than any other CFD graphics program. Other versions available are: SGI IRIS, UNIX Supercomputer and SGI IRIS, DEC VAX using DISSPLA, Generic UNIX using DISSPLA, and Apollo UNIX using GMR3D.

Buning, Pieter G.↗

Research in Parallel Algorithms and Software for Computational Aerosciences

Phase 1 is complete for the development of a computational fluid dynamics CFD) parallel code with automatic grid generation and adaptation for the Euler analysis of flow over complex geometries. SPLITFLOW, an unstructured Cartesian grid code developed at Lockheed Martin Tactical Aircraft Systems, has been modified for a distributed memory/massively parallel computing environment. The parallel code is operational on an SGI network, Cray J90 and C90 vector machines, SGI Power Challenge, and Cray T3D and IBM SP2 massively parallel machines. Parallel Virtual Machine (PVM) is the message passing protocol for portability to various architectures. A domain decomposition technique was developed which enforces dynamic load balancing to improve solution speed and memory requirements. A host/node algorithm distributes the tasks. The solver parallelizes very well, and scales with the number of processors. Partially parallelized and non-parallelized tasks consume most of the wall clock time in a very fine grain environment. Timing comparisons on a Cray C90 demonstrate that Parallel SPLITFLOW runs 2.4 times faster on 8 processors than its non-parallel counterpart autotasked over 8 processors.

Domel, Neal D.↗

Research in Parallel Algorithms and Software for Computational Aerosciences

Phase I is complete for the development of a Computational Fluid Dynamics parallel code with automatic grid generation and adaptation for the Euler analysis of flow over complex geometries. SPLITFLOW, an unstructured Cartesian grid code developed at Lockheed Martin Tactical Aircraft Systems, has been modified for a distributed memory/massively parallel computing environment. The parallel code is operational on an SGI network, Cray J90 and C90 vector machines, SGI Power Challenge, and Cray T3D and IBM SP2 massively parallel machines. Parallel Virtual Machine (PVM) is the message passing protocol for portability to various architectures. A domain decomposition technique was developed which enforces dynamic load balancing to improve solution speed and memory requirements. A host/node algorithm distributes the tasks. The solver parallelizes very well, and scales with the number of processors. Partially parallelized and non-parallelized tasks consume most of the wall clock time in a very fine grain environment. Timing comparisons on a Cray C90 demonstrate that Parallel SPLITFLOW runs 2.4 times faster on 8 processors than its non-parallel counterpart autotasked over 8 processors.

Domel, Neal D.↗

Incorporating the APS Catalog of the POSS I and Image Archive in ADS

The primary purpose of this contract was to develop the software to both create and access an on-line database of images from digital scans of the Palomar Sky Survey. This required modifying our DBMS (called Star Base) to create an image database from the actual raw pixel data from the scans. The digitized images are processed into a set of coordinate-reference index and pixel files that are stored in run-length files, thus achieving an efficient lossless compression. For efficiency and ease of referencing, each digitized POSS I plate is then divided into 900 subplates. Our custom DBMS maps each query into the corresponding POSS plate(s) and subplate(s). All images from the appropriate subplates are retrieved from disk with byte-offsets taken from the index files. These are assembled on-the-fly into a GIF image file for browser display, and a FITS format image file for retrieval. The FITS images have a pixel size of 0.33 arcseconds. The FITS header contains astrometric and photometric information. This method keeps the disk requirements manageable while allowing for future improvements. When complete, the APS Image Database will contain over 130 Gb of data. A set of web pages query forms are available on-line, as well as an on-line tutorial and documentation. The database is distributed to the Internet by a high-speed SGI server and a high-bandwidth disk system. URL is http://aps.umn.edu/IDB/. The image database software is written in perl and C and has been compiled on SGI computers with MIX5.3. A copy of the written documentation is included and the software is on the accompanying exabyte tape.

Humphreys, Roberta M.↗

Building an Integrated Environment for Multimedia

Multimedia courseware on the solar system and earth science suitable for use in elementary, middle, and high schools was developed under this grant. The courseware runs on Silicon Graphics, Incorporated (SGI) workstations and personal computers (PCs). There is also a version of the courseware accessible via the World Wide Web. Accompanying multimedia database systems were also developed to enhance the multimedia courseware. The database systems accompanying the PC software are based on the relational model, while the database systems accompanying the SGI software are based on the object-oriented model.

Source record↗

Virtual Reality Simulation of the International Space Welding Experiment

Virtual Reality (VR) is a set of breakthrough technologies that allow a human being to enter and fully experience a 3-dimensional, computer simulated environment. A true virtual reality experience meets three criteria: (1) It involves 3-dimensional computer graphics; (2) It includes real-time feedback and response to user actions; and (3) It must provide a sense of immersion. Good examples of a virtual reality simulator are the flight simulators used by all branches of the military to train pilots for combat in high performance jet fighters. The fidelity of such simulators is extremely high -- but so is the price tag, typically millions of dollars. Virtual reality teaching and training methods are manifestly effective, and we have therefore implemented a VR trainer for the International Space Welding Experiment. My role in the development of the ISWE trainer consisted of the following: (1) created texture-mapped models of the ISWE's rotating sample drum, technology block, tool stowage assembly, sliding foot restraint, and control panel; (2) developed C code for control panel button selection and rotation of the sample drum; (3) In collaboration with Tim Clark (Antares Virtual Reality Systems), developed a serial interface box for the PC and the SGI Indigo so that external control devices, similar to ones actually used on the ISWE, could be used to control virtual objects in the ISWE simulation; (4) In collaboration with Peter Wang (SFFP) and Mark Blasingame (Boeing), established the interference characteristics of the VIM 1000 head-mounted-display and tested software filters to correct the problem; (5) In collaboration with Peter Wang and Mark Blasingame, established software and procedures for interfacing the VPL DataGlove and the Polhemus 6DOF position sensors to the SGI Indigo serial ports. The majority of the ISWE modeling effort was conducted on a PC-based VR Workstation, described below.

Phillips, James A.↗

3D Multistage Simulation of Each Component of the GE90 Turbofan Engine

A 3D multistage simulation of each component of the GE90 Turbofan engine has been made. This includes 49 blade rows. A coupled simulation of all blade rows will be made very soon. The simulation is running using two levels of parallelism. The first level is on a blade row basis with information shared using files. The second level is using a grid domain decomposition with information shared using MPI. Timings will be shown for running on the SP2, an SGI Origin and a distributed system of HP workstations. On the HP workstations, the CHIMP version of MPI is used, with queuing supplied by LSF (Load Sharing Facility). A script-based control system is used to ensure reliability. An MPEG movie illustrating the flow simulation of the engine has been created using PV3, a parallel visualization library created by Bob Haimes of MIT. PVM is used to create a virtual machine from 10 HP workstations and display on an SGI workstation. A representative component simulation will be compared to rig data to demonstrate its usefulness in turbomachinery design and analysis.

Turner, Mark↗

Aerodynamic Shape Optimization Using A Combined Distributed/Shared Memory Paradigm

Current parallel computational approaches involve distributed and shared memory paradigms. In the distributed memory paradigm, each processor has its own independent memory. Message passing typically uses a function library such as MPI or PVM. In the shared memory paradigm, such as that used on the SGI Origin 2000 machine, compiler directives are used to instruct the compiler to schedule multiple threads to perform calculations. In this paradigm, it must be assured that processors (threads) do not simultaneously access regions of memory in such away that errors would occur. This paper utilizes the latest version of the SGI MPI function library to combine the two parallelization paradigms to perform aerodynamic shape optimization of a generic wing/body.

Cheung, Samson↗

Large-Scale Parallel Viscous Flow Computations using an Unstructured Multigrid Algorithm

The development and testing of a parallel unstructured agglomeration multigrid algorithm for steady-state aerodynamic flows is discussed. The agglomeration multigrid strategy uses a graph algorithm to construct the coarse multigrid levels from the given fine grid, similar to an algebraic multigrid approach, but operates directly on the non-linear system using the FAS (Full Approximation Scheme) approach. The scalability and convergence rate of the multigrid algorithm are examined on the SGI Origin 2000 and the Cray T3E. An argument is given which indicates that the asymptotic scalability of the multigrid algorithm should be similar to that of its underlying single grid smoothing scheme. For medium size problems involving several million grid points, near perfect scalability is obtained for the single grid algorithm, while only a slight drop-off in parallel efficiency is observed for the multigrid V- and W-cycles, using up to 128 processors on the SGI Origin 2000, and up to 512 processors on the Cray T3E. For a large problem using 25 million grid points, good scalability is observed for the multigrid algorithm using up to 1450 processors on a Cray T3E, even when the coarsest grid level contains fewer points than the total number of processors.

Mavriplis, Dimitri J.↗

Advanced Methods for Aircraft Engine Thrust and Noise Benefits: Nozzle-Inlet Flow Analysis

The Fluid Mechanics and Acoustics Laboratory at Hampton University (HU/FM&AL) jointly with the NASA Glenn Research Center has conducted four connected subprojects under the reporting project. Basically, the HU/FM&AL Team has been involved in joint research with the purpose of theoretical explanation of experimental facts and creation of accurate numerical simulation techniques and prediction theory for solution of current problems in propulsion systems of interest to the NAVY and NASA agencies. This work is also supported by joint research between the NASA GRC and the Institute of Mechanics at Moscow State University (IM/MSU) in Russia under a CRDF grant. The research is focused on a wide regime of problems in the propulsion field as well as in experimental testing and theoretical and numerical simulation analyses for advanced aircraft and rocket engines. The FM&AL Team uses analytical methods, numerical simulations and possible experimental tests at the Hampton University campus. The fundamental idea uniting these subprojects is to use nontraditional 3D corrugated and composite nozzle and inlet designs and additional methods for exhaust jet noise reduction without essential thrust loss and even with thrust augmentation. These subprojects are: (1) Aeroperformance and acoustics of Bluebell-shaped and Telescope-shaped designs; (2) An analysis of sharp-edged nozzle exit designs for effective fuel injection into the flow stream in air-breathing engines: triangular-round, diamond-round and other nozzles; (3) Measurement technique improvement for the HU Low Speed Wind Tunnel; a new course in the field of aerodynamics, teaching and training of HU students; experimental tests of Mobius-shaped screws: research and training; (4) Supersonic inlet shape optimization. The main outcomes during this reporting period are: (l) Publications: The AIAA Paper #00-3170 was presented at the 36th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, 17-19 June, 2000, Huntsville, AL. The AIAA Paper #01-1893 has been accepted for the AIAA/NAL-NASDA-ISAS 10th International Space Planes and Hypersonic Systems and Technologies Conference, 24-27 April 2001, Kyoto, Japan. The AIAA Paper #01 -3204 has been accepted for presentation at the 37th AIAA/ASME/SAE/ASEE Joint Propulsion Conference, being held on 08-11 July, in Salt Lake City, UT; (2) A U.S. patent #6,082,635 was granted on July 4, 2000; (3) Grants and proposals: The H U/ FM&AL was awarded the NASA grant NAG-3-2495 in October 2000 and the laboratory is a primary U.S. research team in a joint project under the CRDF award granted to the NASA GRC and IM/MSU (Russia) in July 2000; (4) Theory and numerical simulations: Analytical theory, numerical simulation, comparison of theoretical with experimental results, and modification of theoretical approaches, models, grids, etc., have been conducted for several complicated 2D and 3D nozzle and inlet designs using NASA, ICASE, and IM/MSU codes based on full Euler and Navier-Stokes solvers: CFL3D, FLUENT, and GODUNOV, and others; (5) Experimental Tests: (a) A new course: "Advanced Aerodynamics and Aircraft Performance" presented in spring semester, 2001; training and experimental test research using the HU LSWT. (b) Small-scale M6bius-shaped screws were tested in different conditions and their application has shown essential benefits by comparison with traditional designs; (6) Installation in the FM&AL computer system: second software TECPLOT 8.0 for the UNIX SGI workstation and free TECPLOT 7.5 for the PC Dell computer, and 2D and 3D GRIDGEN (version 9) for the UNIX SGI as well as installation of two free NASA codes, 3D MAG and VULCAN; (7) Student Research Activity: Involvement of two undergraduate students as research assistants in the current research project.

Gilinsky, Mikhail↗

NAS Parallel Benchmark Results 11-96

The NAS Parallel Benchmarks have been developed at NASA Ames Research Center to study the performance of parallel supercomputers. The eight benchmark problems are specified in a "pencil and paper" fashion. In other words, the complete details of the problem to be solved are given in a technical document, and except for a few restrictions, benchmarkers are free to select the language constructs and implementation techniques best suited for a particular system. These results represent the best results that have been reported to us by the vendors for the specific 3 systems listed. In this report, we present new NPB (Version 1.0) performance results for the following systems: DEC Alpha Server 8400 5/440, Fujitsu VPP Series (VX, VPP300, and VPP700), HP/Convex Exemplar SPP2000, IBM RS/6000 SP P2SC node (120 MHz), NEC SX-4/32, SGI/CRAY T3E, SGI Origin200, and SGI Origin2000. We also report High Performance Fortran (HPF) based NPB results for IBM SP2 Wide Nodes, HP/Convex Exemplar SPP2000, and SGI/CRAY T3D. These results have been submitted by Applied Parallel Research (APR) and Portland Group Inc. (PGI). We also present sustained performance per dollar for Class B LU, SP and BT benchmarks.

Bailey, David H.↗

A Multi-Level Parallelization Concept for High-Fidelity Multi-Block Solvers

The integration of high-fidelity Computational Fluid Dynamics (CFD) analysis tools with the industrial design process benefits greatly from the robust implementations that are transportable across a wide range of computer architectures. In the present work, a hybrid domain-decomposition and parallelization concept was developed and implemented into the widely-used NASA multi-block Computational Fluid Dynamics (CFD) packages implemented in ENSAERO and OVERFLOW. The new parallel solver concept, PENS (Parallel Euler Navier-Stokes Solver), employs both fine and coarse granularity in data partitioning as well as data coalescing to obtain the desired load-balance characteristics on the available computer platforms. This multi-level parallelism implementation itself introduces no changes to the numerical results, hence the original fidelity of the packages are identically preserved. The present implementation uses the Message Passing Interface (MPI) library for interprocessor message passing and memory accessing. By choosing an appropriate combination of the available partitioning and coalescing capabilities only during the execution stage, the PENS solver becomes adaptable to different computer architectures from shared-memory to distributed-memory platforms with varying degrees of parallelism. The PENS implementation on the IBM SP2 distributed memory environment at the NASA Ames Research Center obtains 85 percent scalable parallel performance using fine-grain partitioning of single-block CFD domains using up to 128 wide computational nodes. Multi-block CFD simulations of complete aircraft simulations achieve 75 percent perfect load-balanced executions using data coalescing and the two levels of parallelism. SGI PowerChallenge, SGI Origin 2000, and a cluster of workstations are the other platforms where the robustness of the implementation is tested. The performance behavior on the other computer platforms with a variety of realistic problems will be included as this on-going study progresses.

Hatay, Ferhat F.↗

Video Editing System

This is a proposal for a general use system based, on the SGI IRIS workstation platform, for recording computer animation to videotape. In addition, this system would provide features for simple editing and enhancement. Described here are a list of requirements for the system, and a proposed configuration including the SGI VideoLab Integrator, VideoMedia VLAN animation controller and the Pioneer rewritable laserdisc recorder.

Schlecht, Leslie E.↗