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 19 records

Biochemical changes of salivary gland adenoid cystic carcinoma cells induced by SGI-1776

Provirus integration site for Moloney murine leukemia virus 1 (Pim-1) has proved to be an oncogene and it is known that to depress Pim-1 activity may be a novel oncological treatment strategy. SGI-1776, a small molecule, is the first clinically tested inhibitor of the Pim kinase family. Here, we aimed to explore the effect of SGI-1776 on salivary adenoid cystic carcinoma (SACC). Expression of Pim-1 was confirmed in SACC and control tissues by qRT-PCR. After SGI-1776 treatment, the Pim-1 expressions and Pim-1 kinase activity in both SACC-83 and SACC-LM cell lines were measured. Cell proliferation, cell invasion, cell cycle, apoptosis and mitochondrial membrane potential were analyzed. Also, the expression of FOXO3a, p-FOXO3a, RUNX3, Bcl-2, BAD, p-BAD, Bim and p-Bim were detected by Western blot. The results showed that Pim-1 was significantly overexpressed in SACC tissues. SGI-1776 down-regulated the Pim-1 expression, inhibited Pim-1 kinase activity, reduced cell proliferation, decreased invasive ability, increased caspase-3 activity and induced apoptosis, cell cycle arrest and mitochondrial depolarization. Reduced expression was also seen in p-FOXO3a, RUNX3, Bcl-2, p-BAD and p-Bim, whereas no significant changes were observed from FOXO3a, BAD and Bim. These results confirm the pivotal role of Pim-1 in SACC and suggest that targeting Pim-1 kinase signal pathway by SGI-1776 might be a promising therapeutic modality for SACC.

60 APPLIED LIFE SCIENCES↗

10 Gigabit Ethernet Performance on SGI Altix and Origin Systems

As the state of high performance computing continues to advance, the size of datasets continue to grow, driving a need for high bandwidth data networks. family of networks. 10 Gigabit Ethernet is the latest step in the popular Ethernet We have evaluated the S2io Xframe 10 Gigabit Ethernet adapter on 512p SGI Altix systems running ProPack 3, and Origin systems running Irix 6.5.24 and 6.5.26 in our production supercomputing environment. We encountered a number of performance and stability issues, which were promptly dealt with by SGI and S2io. Using nttcp we tested TCP performance for single and multiple streams, and we tested file transfer using NFS and bbftp. We will present the results of our testing, including the effects of various tuning options on throughput and CPU utilization, and offer suggestions for configuring and tuning S2io 10 Gigabit Ethernet cards in an Altix/Linux or Origin/Irix environment.

Meyer, Andy↗

SGI Implementation of MPI-2

MPI-2 is the natural successor to MPI-1 in many ways, and includes 3 major features: parallel I/O (which has a NAS legacy) , remote memory operations (alias one-sided I/O), and dynamic process management. SGI supports a subset of MPI-2. It is useful to have a map of what one can and cannot do with MPI-2 on SGI Origin and Altix systems, to be aware of known problems, and to contrast performance on different platforms.

Nelson, Terry↗

NAS Experiences of Porting CM Fortran Codes to HPF on IBM SP2 and SGI Power Challenge

Current Connection Machine (CM) Fortran codes developed for the CM-2 and the CM-5 represent an important class of parallel applications. Several users have employed CM Fortran codes in production mode on the CM-2 and the CM-5 for the last five to six years, constituting a heavy investment in terms of cost and time. With Thinking Machines Corporation's decision to withdraw from the hardware business and with the decommissioning of many CM-2 and CM-5 machines, the best way to protect the substantial investment in CM Fortran codes is to port the codes to High Performance Fortran (HPF) on highly parallel systems. HPF is very similar to CM Fortran and thus represents a natural transition. Conversion issues involved in porting CM Fortran codes on the CM-5 to HPF are presented. In particular, the differences between data distribution directives and the CM Fortran Utility Routines Library, as well as the equivalent functionality in the HPF Library are discussed. Several CM Fortran codes (Cannon algorithm for matrix-matrix multiplication, Linear solver Ax=b, 1-D convolution for 2-D datasets, Laplace's Equation solver, and Direct Simulation Monte Carlo (DSMC) codes have been ported to Subset HPF on the IBM SP2 and the SGI Power Challenge. Speedup ratios versus number of processors for the Linear solver and DSMC code are presented.

Saini, Subhash↗

A Comparison of Automatic Parallelization Tools/Compilers on the SGI Origin 2000 Using the NAS Benchmarks

Porting applications to new high performance parallel and distributed computing platforms is a challenging task. Since writing parallel code by hand is extremely time consuming and costly, porting codes would ideally be automated by using some parallelization tools and compilers. In this paper, we compare the performance of the hand written NAB Parallel Benchmarks against three parallel versions generated with the help of tools and compilers: 1) CAPTools: an interactive computer aided parallelization too] that generates message passing code, 2) the Portland Group's HPF compiler and 3) using compiler directives with the native FORTAN77 compiler on the SGI Origin2000.

Saini, Subhash↗

Interconnect Performance Evaluation of SGI Altix 3700 BX2, Cray X1, Cray Opteron Cluster, and Dell PowerEdge

We study the performance of inter-process communication on four high-speed multiprocessor systems using a set of communication benchmarks. The goal is to identify certain limiting factors and bottlenecks with the interconnect of these systems as well as to compare these interconnects. We measured network bandwidth using different number of communicating processors and communication patterns, such as point-to-point communication, collective communication, and dense communication patterns. The four platforms are: a 512-processor SGI Altix 3700 BX2 shared-memory machine with 3.2 GB/s links; a 64-processor (single-streaming) Cray XI shared-memory machine with 32 1.6 GB/s links; a 128-processor Cray Opteron cluster using a Myrinet network; and a 1280-node Dell PowerEdge cluster with an InfiniBand network. Our, results show the impact of the network bandwidth and topology on the overall performance of each interconnect.

Fatoohi, Rod↗

Supersonic Gas Injector for Plasma Fueling in the National Spherical Torus Experiment

A supersonic gas injector (SGI) has been developed for fueling and diagnostic applications on the National Spherical Torus Experiment (NSTX). It is comprised of a graphite converging-diverging Laval nozzle and a commercial piezoelectric gas valve mounted on a movable probe at a low-field-side midplane port location. Also mounted on the probe is a diagnostic package: a Langmuir probe, two thermocouples, and five pick-up coils for measuring toroidal, radial, vertical magnetic field components and magnetic fluctuations at the location of the SGI tip. The SGI flow rate is up to 33.25 Pa m 3 / (1.75 × 10 22 euterium particles/s), comparable to conventional NSTX gas injectors. The nozzle operates in a pulsed regime at room temperature and a reservoir gas pressure up to 665 kPa (5000 Torr). The deuterium jet Mach number of about 4 and the divergence half-angle of 5 to 25 deg have been measured in laboratory experiments simulating the NSTX environment. Reliable operation of the SGI and all mounted diagnostics at distances 0.01 to 0.20 m from the plasma separatrix has been demonstrated in NSTX experiments. The SGI has been used for fueling of ohmic and 2- to 4-MW neutral beam injection–heated L- and H-mode plasmas. Fueling efficiency in the range 0.1 to 0.3 has been obtained from the plasma electron inventory analysis. In conclusion, the SGI-fueling–based plasma discharge scenarios enabling better density control have been developed.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

H-mode fueling optimization with the supersonic deuterium jet in NSTX

High-performance, long-pulse 0.7-1.2 MA 6-7 MW NBI-heated small-ELM H-mode plasma discharges are developed in the National Spherical Torus Experiment (NSTX) as prototypes for confinement and current drive extrapolations to future spherical tori. It is envisioned that innovative lithium coating techniques for H-mode density pumping and a supersonic deuterium jet for plasma refueling will be used to achieve the low pedestal collisionality and low n{sub e}/n{sub G} fractions (0.3-0.6), both of which being essential conditions for maximizing the non-inductive (bootstrap and beam driven) current fractions. The low field side supersonic gas injector (SGI) on NSTX consists of a small converging-diverging graphite Laval nozzle and a piezoelectric gas valve. The nozzle is capable of producing a deuterium jet with Mach number M {le} 4, estimated gas density at the nozzle exit n {le} 5 x 10{sup 23} m{sup -3}, estimated temperature T {ge} 70 K, and flow velocity v = 2:4 km/s. The nozzle Reynolds number Reis {approx_equal} 6000. The nozzle and the valve are enclosed in a protective carbon fiber composite shroud and mounted on a movable probe at a midplane port location. Despite the beneficial L-mode fueling experience with supersonic jets in limiter tokamaks, there is a limited experience with fueling of high-performance H-mode divertor discharges and the associated density, MHD stability, and MARFE limits. In initial supersonic deuterium jet fueling experiments in NSTX, a reliable H-mode access, a low NBI power threshold, P{sub LH} {le} 2 MW, and a high fueling efficiency (0.1-0.4) have been demonstrated. Progress has also been made toward a better control of the injected fueling gas by decreasing the uncontrolled high field side (HFS) injector fueling rate by up to 95 % and complementing it with the supersonic jet fueling. These results motivated recent upgrades to the SGI gas delivery and control systems. The new SGI-Upgrade (SGI-U) capabilities include multi-pulse ms-scale controls and a reservoir gas pressure up to P{sub 0} = 5000 Torr. In this paper we summarize recent progress toward optimization of H-mode fueling in NSTX using the SGI-U.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Engineering Nucleation Kinetics of Graphite Nodules in Inoculated Cast Iron for Reducing Porosity

Inoculation of cast iron with spherical graphite (SGI) controls nucleation kinetics of graphite nodules and is used for improving casting performance and eliminating solidification micro-porosity. In this study, thermodynamic simulations were used to design inoculants. Thermal and chemical stability of different potential heterogeneous nuclei formed in the melt above the liquidus temperature and in the mushy zone during solidification was predicted. Several inoculation treatments of SGI were performed in laboratory heats. An automated SEM/EDX analysis was applied to examine the graphite nodule size distribution and the family of the non-metallic inclusions in the experimental castings. The data were used to reconstruct a relative graphite nodule nucleation rate in the castings. It was shown that the graphite nodule nucleation kinetics in the castings were significantly different from those predicted by classical nucleation models. In inoculated SGI, the observed bi-modal distribution of graphite nodules was related to a continuous nucleation that occurred towards the end of solidification. The origin of the continuous nucleation and the possibility of engineering graphite nodules nucleation kinetics to control SGI casting soundness by inoculation are discussed.

36 MATERIALS SCIENCE↗

Relationship and distribution of Salmonella enterica serovar I 4,[5],12:i:- strain sequences in the NCBI Pathogen Detection database

Background: Of the > 2600 Salmonella serovars, Salmonella enterica serovar I 4,[5],12:i:- (serovar I 4,[5],12:i:-) has emerged as one of the most common causes of human salmonellosis and the most frequent multidrug-resistant (MDR; resistance to ≥3 antimicrobial classes) nontyphoidal Salmonella serovar in the U.S. Serovar I 4,[5],12:i:- isolates have been described globally with resistance to ampicillin, streptomycin, sulfisoxazole, and tetracycline (R-type ASSuT) and an integrative and conjugative element with multi-metal tolerance named Salmonella Genomic Island 4 (SGI-4). Results: We analyzed 13,612 serovar I 4,[5],12:i:- strain sequences available in the NCBI Pathogen Detection database to determine global distribution, animal sources, presence of SGI-4, occurrence of R-type ASSuT, frequency of antimicrobial resistance (AMR), and potential transmission clusters. Genome sequences for serovar I 4,[5],12:i:- strains represented 30 countries from 5 continents (North America, Europe, Asia, Oceania, and South America), but sequences from the United States (59%) and the United Kingdom (28%) were dominant. The metal tolerance island SGI-4 and the R-type ASSuT were present in 71 and 55% of serovar I 4,[5],12:i:- strain sequences, respectively. Sixty-five percent of strain sequences were MDR which correlates to serovar I 4,[5],12:i:- being the most frequent MDR serovar. The distribution of serovar I 4,[5],12:i:- strain sequences in the NCBI Pathogen Detection database suggests that swine-associated strain sequences were the most frequent food-animal source and were significantly more likely to contain the metal tolerance island SGI-4 and genes for MDR compared to all other animal-associated isolate sequences. Conclusions: Our study illustrates how analysis of genomic sequences from the NCBI Pathogen Detection database can be utilized to identify the prevalence of genetic features such as antimicrobial resistance, metal tolerance, and virulence genes that may be responsible for the successful emergence of bacterial foodborne pathogens.

59 BASIC BIOLOGICAL SCIENCES↗

Parallelization of Lower-Upper Symmetric Gauss-Seidel Method for Chemically Reacting Flow

Development of technologies for exploration of the solar system has revived an interest in computational simulation of chemically reacting flows since planetary probe vehicles exhibit non-equilibrium phenomena during the atmospheric entry of a planet or a moon as well as the reentry to the Earth. Stability in combustion is essential for new propulsion systems. Numerical solution of real-gas flows often increases computational work by an order-of-magnitude compared to perfect gas flow partly because of the increased complexity of equations to solve. Recently, as part of Project Columbia, NASA has integrated a cluster of interconnected SGI Altix systems to provide a ten-fold increase in current supercomputing capacity that includes an SGI Origin system. Both the new and existing machines are based on cache coherent non-uniform memory access architecture. Lower-Upper Symmetric Gauss-Seidel (LU-SGS) relaxation method has been implemented into both perfect and real gas flow codes including Real-Gas Aerodynamic Simulator (RGAS). However, the vectorized RGAS code runs inefficiently on cache-based shared-memory machines such as SGI system. Parallelization of a Gauss-Seidel method is nontrivial due to its sequential nature. The LU-SGS method has been vectorized on an oblique plane in INS3D-LU code that has been one of the base codes for NAS Parallel benchmarks. The oblique plane has been called a hyperplane by computer scientists. It is straightforward to parallelize a Gauss-Seidel method by partitioning the hyperplanes once they are formed. Another way of parallelization is to schedule processors like a pipeline using software. Both hyperplane and pipeline methods have been implemented using openMP directives. The present paper reports the performance of the parallelized RGAS code on SGI Origin and Altix systems.

Yoon, Seokkwan↗

LARGE-SCALE FILAMENTARY STRUCTURES AROUND THE VIRGO CLUSTER REVISITED

We revisit the filamentary structures of galaxies around the Virgo cluster, exploiting a larger data set, based on the HyperLeda database, than previous studies. In particular, this includes a large number of low-luminosity galaxies, resulting in better sampled individual structures. We confirm seven known structures in the distance range 4 h {sup −1} Mpc < SGY < 16 h {sup −1} Mpc, now identified as filaments, where SGY is the axis of the supergalactic coordinate system roughly along the line of sight. The Hubble diagram of the filament galaxies suggests they are infalling toward the main body of the Virgo cluster. We propose that the collinear distribution of giant elliptical galaxies along the fundamental axis of the Virgo cluster is smoothly connected to two of these filaments (Leo II A and B). Behind the Virgo cluster (16 h {sup −1} Mpc < SGY < 27 h {sup −1} Mpc), we also identify a new filament elongated toward the NGC 5353/4 group (“NGC 5353/4 filament”) and confirm a sheet that includes galaxies from the W and M clouds of the Virgo cluster (“W–M sheet”). In the Hubble diagram, the NGC 5353/4 filament galaxies show infall toward the NGC 5353/4 group, whereas the W–M sheet galaxies do not show hints of gravitational influence from the Virgo cluster. The filamentary structures identified can now be used to better understand the generic role of filaments in the build-up of galaxy clusters at z ≈ 0.

79 ASTRONOMY AND ASTROPHYSICS↗

Parallel Implementation of the Discontinuous Galerkin Method

This paper describes a parallel implementation of the discontinuous Galerkin method. Discontinuous Galerkin is a spatially compact method that retains its accuracy and robustness on non-smooth unstructured grids and is well suited for time dependent simulations. Several parallelization approaches are studied and evaluated. The most natural and symmetric of the approaches has been implemented in all object-oriented code used to simulate aeroacoustic scattering. The parallel implementation is MPI-based and has been tested on various parallel platforms such as the SGI Origin, IBM SP2, and clusters of SGI and Sun workstations. The scalability results presented for the SGI Origin show slightly superlinear speedup on a fixed-size problem due to cache effects.

Baggag, Abdalkader↗

Use Computer-Aided Tools to Parallelize Large CFD Applications

Porting applications to high performance parallel computers is always a challenging task. It is time consuming and costly. With rapid progressing in hardware architectures and increasing complexity of real applications in recent years, the problem becomes even more sever. Today, scalability and high performance are mostly involving handwritten parallel programs using message-passing libraries (e.g. MPI). However, this process is very difficult and often error-prone. The recent reemergence of shared memory parallel (SMP) architectures, such as the cache coherent Non-Uniform Memory Access (ccNUMA) architecture used in the SGI Origin 2000, show good prospects for scaling beyond hundreds of processors. Programming on an SMP is simplified by working in a globally accessible address space. The user can supply compiler directives, such as OpenMP, to parallelize the code. As an industry standard for portable implementation of parallel programs for SMPs, OpenMP is a set of compiler directives and callable runtime library routines that extend Fortran, C and C++ to express shared memory parallelism. It promises an incremental path for parallel conversion of existing software, as well as scalability and performance for a complete rewrite or an entirely new development. Perhaps the main disadvantage of programming with directives is that inserted directives may not necessarily enhance performance. In the worst cases, it can create erroneous results. While vendors have provided tools to perform error-checking and profiling, automation in directive insertion is very limited and often failed on large programs, primarily due to the lack of a thorough enough data dependence analysis. To overcome the deficiency, we have developed a toolkit, CAPO, to automatically insert OpenMP directives in Fortran programs and apply certain degrees of optimization. CAPO is aimed at taking advantage of detailed inter-procedural dependence analysis provided by CAPTools, developed by the University of Greenwich, to reduce potential errors made by users. Earlier tests on NAS Benchmarks and ARC3D have demonstrated good success of this tool. In this study, we have applied CAPO to parallelize three large applications in the area of computational fluid dynamics (CFD): OVERFLOW, TLNS3D and INS3D. These codes are widely used for solving Navier-Stokes equations with complicated boundary conditions and turbulence model in multiple zones. Each one comprises of from 50K to 1,00k lines of FORTRAN77. As an example, CAPO took 77 hours to complete the data dependence analysis of OVERFLOW on a workstation (SGI, 175MHz, R10K processor). A fair amount of effort was spent on correcting false dependencies due to lack of necessary knowledge during the analysis. Even so, CAPO provides an easy way for user to interact with the parallelization process. The OpenMP version was generated within a day after the analysis was completed. Due to sequential algorithms involved, code sections in TLNS3D and INS3D need to be restructured by hand to produce more efficient parallel codes. An included figure shows preliminary test results of the generated OVERFLOW with several test cases in single zone. The MPI data points for the small test case were taken from a handcoded MPI version. As we can see, CAPO's version has achieved 18 fold speed up on 32 nodes of the SGI O2K. For the small test case, it outperformed the MPI version. These results are very encouraging, but further work is needed. For example, although CAPO attempts to place directives on the outer- most parallel loops in an interprocedural framework, it does not insert directives based on the best manual strategy. In particular, it lacks the support of parallelization at the multi-zone level. Future work will emphasize on the development of methodology to work in a multi-zone level and with a hybrid approach. Development of tools to perform more complicated code transformation is also needed.

Jin, H.↗

A Linux Workstation for High Performance Graphics

The primary goal of this effort was to provide a low-cost method of obtaining high-performance 3-D graphics using an industry standard library (OpenGL) on PC class computers. Previously, users interested in doing substantial visualization or graphical manipulation were constrained to using specialized, custom hardware most often found in computers from Silicon Graphics (SGI). We provided an alternative to expensive SGI hardware by taking advantage of third-party, 3-D graphics accelerators that have now become available at very affordable prices. To make use of this hardware our goal was to provide a free, redistributable, and fully-compatible OpenGL work-alike library so that existing bodies of code could simply be recompiled. for PC class machines running a free version of Unix. This should allow substantial cost savings while greatly expanding the population of people with access to a serious graphics development and viewing environment. This should offer a means for NASA to provide a spectrum of graphics performance to its scientists, supplying high-end specialized SGI hardware for high-performance visualization while fulfilling the requirements of medium and lower performance applications with generic, off-the-shelf components and still maintaining compatibility between the two.

Geist, Robert↗

Compute Server Performance Results

Parallel-vector supercomputers have been the workhorses of high performance computing. As expectations of future computing needs have risen faster than projected vector supercomputer performance, much work has been done investigating the feasibility of using Massively Parallel Processor systems as supercomputers. An even more recent development is the availability of high performance workstations which have the potential, when clustered together, to replace parallel-vector systems. We present a systematic comparison of floating point performance and price-performance for various compute server systems. A suite of highly vectorized programs was run on systems including traditional vector systems such as the Cray C90, and RISC workstations such as the IBM RS/6000 590 and the SGI R8000. The C90 system delivers 460 million floating point operations per second (FLOPS), the highest single processor rate of any vendor. However, if the price-performance ration (PPR) is considered to be most important, then the IBM and SGI processors are superior to the C90 processors. Even without code tuning, the IBM and SGI PPR's of 260 and 220 FLOPS per dollar exceed the C90 PPR of 160 FLOPS per dollar when running our highly vectorized suite,

Stockdale, I. E.↗

Efficacy of Code Optimization on Cache-Based Processors

In this paper a number of techniques for improving the cache performance of a representative piece of numerical software is presented. Target machines are popular processors from several vendors: MIPS R5000 (SGI Indy), MIPS R8000 (SGI PowerChallenge), MIPS R10000 (SGI Origin), DEC Alpha EV4 + EV5 (Cray T3D & T3E), IBM RS6000 (SP Wide-node), Intel PentiumPro (Ames' Whitney), Sun UltraSparc (NERSC's NOW). The optimizations all attempt to increase the locality of memory accesses. But they meet with rather varied and often counterintuitive success on the different computing platforms. We conclude that it may be genuinely impossible to obtain portable performance on the current generation of cache-based machines. At the least, it appears that the performance of modern commodity processors cannot be described with parameters defining the cache alone.

VanderWijngaart, Rob F.↗