Search NASA⌕ Search

SEARCH · Search NASA

Results for “Overflow”

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 307 records · Page 17

NCV Flow Diagnostic Test Results

There were two objectives for this test. First, was to assess the reasons why there is approximately 1.5 drag counts (cts) discrepancy between measured and computed drag improvement of the Non-linear Cruise Validation (NCV) over the Technology Concept Airplane (TCA) wing body (WB) configurations. The Navier-Stokes (N-S) pre-test predictions from Boeing Commercial Airplane Group (BCAG) show 4.5 drag cts of improvement for NCV over TCA at a lift coefficient (CL) of 0. I at Mach 2.4. The pre-test predictions from Boeing Phantom Works - Long Beach, BPW-LB, show 3.75 drag cts of improvement. BCAG used OVERFLOW and BPW-LB used CFL3D. The first test entry to validate the improvement was held at the NASA Langley Research Center (LARC) UPV;T, test number 1687. The experimental results showed that the drag improvement was only 2.6 cts, not accounting for laminar run and trip drag. This is approximately 1.5 cts less than predicted computationally. In addition to the low Reynolds Number (RN) test, there was a high RN test in the Boeing Supersonic Wind Tunnel (BSWT) of NCV and TCA. BSV@T test 647 showed that the drag improvement of NCV over TCA was also 2.6 cts, but this did account for laminar run and trip drag. Every effort needed to be done to assess if the improvement measured in LaRC UPWT and BSWT was correct. The second objective, once the first objective was met, was to assess the performance increment of NCV over TCA accounting for the associated laminar run and trip drag corrections in LaRC UPWT. We know that the configurations tested have laminar flow on portions of the wing and have trip drag due to the mechanisms used to force the flow to go from laminar to turbulent aft of the transition location.

Cappuccio, Mina↗

Skin Friction Drag Predictions: Summary of CFD Cross Checks, Wing/Body

The objective of the present study was to address the questions of: 1) how reliably or consistently the Navier-Stokes methods and processes used by the various organizations can predict integrated skin friction drag, and 2) how well the methods can predict trends within a family of optimized configurations. As a first step, all available skin friction drag predictions were accumulated to obtain a mean and standard deviation for the TCA (Technology Concept Airplane) baseline and each of the optimized configurations. It is observed that the optimization process has had little effect on the predicted skin friction drags. The variation in the mean that is observed is dwarfed by the standard deviations. In order to understand the reasons for the relatively large spreads in the computed results, a number of auxiliary computations have been performed using the UPS and OVERFLOW codes in an effort to identify and quantity potential sources of the variations.

Lawrence, Scott↗

HSCT Propulsion Airframe Integration Studies

The Lockheed Martin spillage study was a substantial effort and is worthy of a separate paper. However, since a paper was not submitted a few of the most pertinent results have been pulled out and included in this paper. The reader is urged to obtain a copy of the complete Boeing Configuration Aerodynamics final 1995 contract report for the complete Lockheed documentation of the spillage work. The supersonic cruise studies presented here focus on the bifurcated - axisymmetric inlet drag delta. In the process of analyzing this delta several test/CFD data correlation problems arose that lead to a correction of the measured drag delta from 4.6 counts to 3.1 counts. This study also lead to much better understanding of the OVERFLOW gridding and solution process, and to increased accuracy of the force and moment data. Detailed observations of the CFD results lead to the conclusion that the 3.1 count difference between the two inlet types could be reduced to approximately 2 counts, with an absolute lower bound of 1.2 counts due to friction drag and the bifurcated lip bevel.

Chaney, Steve↗

HSCT Ref-H Transonic Flap Data Base: Wind-Tunnel Test and Comparison with Theory

In cooperation with personnel from the Boeing ANP Laboratory and NASA Langley, a performance test was conducted using the Reference-H 1.675% model ("NASA Modular Model") without nacelles at the NASA Langley 16-Ft Transonic Tunnel. The main objective of the test was to determine the drag reduction achievable with leading-edge and trailing-edge flaps deflected along the outboard wing span at transonic Mach numbers (M = 0.9 to 1.2) for purpose of preliminary design and for comparison with computational predictions. The obtained drag data with flap deflections for Mach numbers of 1.07 to 1.20 are unique for the Reference H wing. Four leading-edge and two trailing-edge flap deflection angles were tested at a mean-wing chord-Reynolds number of about 5.7 million. An outboard-wing leading-edge flap deflection of 81 provides a 4.5 percent drag reduction at M = 1.2 A = 0.2), and much larger values at lower Mach numbers with larger flap deflections. The present results for the baseline (no flaps deflected) compare reasonably well with previous Boeing and NASA Ref-H tunnel tests, including high-Reynolds number NTF results. Viscous CFD simulations using the OVERFLOW thin-layer N.S. method properly predict the observed trend in drag reduction at M = 1.2 as function of leading-edge flap deflection. Modified linear theory properly predicts the flap effects on drag at subsonic conditions (Aero2S code), and properly predicts the absolute drag for the 40 and 80 leading-edge deflection at M = 1.2 (A389 code).

Vijgen, Paul M.↗

Analysis and Multipoint Design of the TCA Concept

The goal in this effort is to analyze the baseline TCA concept at transonic and supersonic cruise, then apply the natural flow wing design concept to obtain multipoint performance improvements. Analyses are conducted with OVERFLOW, a Navier-Stokes code for overset grids, using PEGSUS to compute the interpolations between the overset grids.

Krist, Steven E.↗

Tubular Coupling

A system for coupling a vascular overflow graft or cannula to a heart pump. A pump pipe outlet is provided with an external tapered surface which receives the end of a compressible connula. An annular compression ring with a tapered internal bore surface is arranged about the cannula with the tapered internal surface in a facing relationship to the external tapered surface. The angle of inclination of the tapered surfaces is converging such that the spacing between the tapered surfaces decreases from one end of the external tapered surface to the other end thereby providing a clamping action of the tapered surface on a cannula which increases as a function of the length of cannula segment between the tapered surfaces. The annular compression ring is disposed within a tubular locking nut which threadedly couples to the pump and provides a compression force for urging the annular ring onto the cannula between the tapered surfaces. The nut has a threaded connection to the pump body. The threaded coupling to the pump body provides a compression force for the annular ring. The annular ring has an annular enclosure space in which excess cannula material from the compression between the tapered surfaces to "bunch up" in the space and serve as an enlarged annular ring segment to assist holding the cannula in place. The clamped cannula provides a seamless joint connection to the pump pipe outlet where the clamping force is uniformly applied to the cannula because of self alignment of the tapered surfaces. The nut can be easily disconnected to replace the pump if necessary.

Rosenbaum, Bernard J.↗

Analysis and Multipoint Design of the TCA Concept

The goal in this effort is to analyze the baseline TCA concept at transonic and supersonic cruise, then apply the natural flow wing design concept to obtain multipoint performance improvements. Analyses are conducted with OVERFLOW, a Navier-Stokes code for overset grids, using PEGSUS to compute the interpolations between the overset grids.

Krist, Steven E.↗

Micromechanical Characterization and Texture Analysis of Direct Cast Titanium Alloys Strips

This research was conducted to determine a post-processing technique to optimize mechanical and material properties of a number of Titanium based alloys and aluminides processed via Melt Overflow Solidification Technique (MORST). This technique was developed by NASA for the development of thin sheet titanium and titanium aluminides used in high temperature applications. The materials investigated in this study included conventional titanium alloy strips and foils, Ti-1100, Ti-24Al-11Nb (Alpha-2), and Ti-48Al-2Ta (Gamma). The methodology used included micro-characterization, heat-treatment, mechanical processing and mechanical testing. Characterization techniques included optical, electron microscopy, and x-ray texture analysis. The processing included heat-treatment and mechanical deformation through cold rolling. The initial as-cast materials were evaluated for their microstructure and mechanical properties. Different heat-treatment and rolling steps were chosen to process these materials. The properties were evaluated further and a processing relationship was established in order to obtain an optimum processing condition. The results showed that the as-cast material exhibited a Widmanstatten (fine grain) microstructure that developed into a microstructure with larger grains through processing steps. The texture intensity showed little change for all processing performed in this investigation.

Source record↗

Proposed Mars Surveyor Landing Sites in Northern Meridiani Sinus, Southern Elysium Planitia, and Argyre Planitia

Our objective is to propose two landing sites that the Mars Surveyor 2001 Lander and Athena Rover could go to on Mars that should meet the safety requirements of the spacecraft landing system and optimize surface operations (chiefly driven by power and communications requirements). An additional site within Argyre Planitia, initially proposed by Parker to the Mars Surveyor Landing Site program, is also proposed for potential consideration for post-2001 missions to Mars, as it is well outside the current latitude limits for the Athena Rover. All three sites are designed to be situated as close to a diversity of geologic units within a few kilometers of the landing site so that diversity can be placed in a geologic context. This objective is very different from the Mars Pathfinder requirement to land at a site with a maximum chance for containing a diversity of rocks within a few tens of meters of the lander. That requirement was driven by the Sojourner mobility limit of a few tens of meters. It can be argued that the Athena project, with its much larger mobility capability, might actually want to avoid such a site, because placing collected samples in geologic context would be difficult. While it has been argued, both before and after the Mars Pathfinder landing, that the provenance for local blocks may be determined by orbiter spectra, primarily from the MGS TES instrument, our ability to do so has yet to be demonstrated. Indeed, several months after conclusion of the Pathfinder mission, we have yet to reach a consensus on the composition of local materials. Our primary data set for selecting a landing site within the latitude and elevation constraints of the 2001 mission is the Viking Orbiter image archive. The site must be selected to place the landing ellipse so as to avoid obvious hazards, such as steep slopes, large or numerous craters, or abundant large knobs. For this purpose, we chose a resolution limit of better than 50 m/pixel. This necessarily excludes from the present study images from current and future orbiter spacecraft, until such data does become readily available. Within each proposed region, it may be possible to identify additional sites once these data become available. Second, the fine-component thermal inertia data, should be greater than about 5 or 6 cgs Units (10(exp -3) cal/sq cm s(exp -0.5)/K). Low thermal inertias imply dusty environments, which could pose a mobility hazard. Similarly, the albedo of the site should not be particularly high, which would also suggest dusty surfaces. Low albedos are preferred, as they often coincide with low Viking red:violet ratios and indicate less dusty surfaces. Next, the Modeled Block Abundance should also not be too high or too low. Based on the Viking Lander and Mars Pathfinder experiences, percentages of blocks should be on the order of 5-25%. Too many blocks could pose a hazard to the landing and mobility. Too few blocks could also indicate a dusty surface. Primary Landing Site: Northern Meridiani Sinus (Proposed by T. J. Parker and K., S. Edgett) Vital Statistics: (1) Latitude, Longitude: 0-3 N, 350-2 W. *Elevation (Viking): about0.5-1.5 Ian. (2) Viking Orbiter Image coverage: Excellent coverage by 15 - 25 m/pixel images (orbits 709A and 410B). Possible stereo coverage in region where two orbits overlap (probably small parallax angle, as these orbits are not listed in NASA Contractor Report 3501) (3) Albedo: about .18 -.26 (4) Block Abundance: 5-26% (5)Fine-Component Thermal Inertia: 5-9 cgs units This region consists of bright deposits similar to those described by Edgett et al, that also lie within a prominent dark albedo region. These deposits are flat-lying, to such a degree that they ramp against topography rather than draping over it. This led Edgett and Parker to suggest that they may be subaqueous sediments, possibly lacustrine or marine evaporites, laid down sometime from the late Noachian to middle Hesperian (age determination pending crater counts). A contact between this material and elevated, dissected highlands to the south was identified , and is described by Edgett et al. Our desire in proposing this landing site is to sample the edge of this deposit where it has been exposed through etching, presumably eolian deflation (the deposit, though in the highlands, is itself only lightly to moderately cratered). This should enable access to in situ stratigraphy. The actual landing site will be selected where slopes are not expected to be steep, such that the rover itself should be able to traverse them and sample layered materials on the way, either up or down the slope. Perhaps due to uncertainties at this time as to the friability or meter-scale roughness of the deposit, it might make sense to place the landing ellipse on the exhumed highland surface adjacent to the deflated margin of the deposit and plan on driving to the deposit rather than landing on it and driving downslope. This should also enable imaging the margin for evidence of layering should it prove too difficult to climb. A target ellipse on the highland surface should also allow Athena access to ancient Noachian highland materials, particularly if placed near crater ejecta or an inlier of knobby material. Secondary Landing Site: Southern Elysium Planitia (Proposed by T. J. Parker) Vital Statistics: (1) Latitude, Longitude: 1.5-3.5 S, 195-198 W. (2) Elevation (Viking): -1.0 km. (3) Viking Orbiter Image coverage: Excellent coverage by 15 - 25 m/pixel images (orbit 725). Possible stereo coverage between images from beginning and end of orbit that overlap (probably small parallax angle) (4) Albedo: about .27-.28 (5) Block Abundance: 4-7% (6) Fine-Component Thermal Inertia: about 3 cgs units This region consists of eroded knobby material, probably of Noachian age, though much of the crater population has been destroyed, that is onlapped at a sharp contact by an extensive plains unit in southern Elysium Planitia that is Amazonian in age. The plains materials have been attributed to unusually low-viscosity flood lavas from fissures south of the Elysium volcanic rise, or to lacustrine materials associated with a large, Amazonian lake at the source of Marte Vallis. Parker and Schenk presented evidence in support of the latter interpretation, though they attributed the putative shore morphology to an embayment of a northern plains ocean into the southern Elysium region. Detailed examination of the margin of the deposit, showing erosion, not simply burial, of small crater rims and fluidized ejecta blankets, also points to lacustrine or marine sedimentation rather than volcanic plains burial. The plains surface exhibits a "crusty" appearance that many researchers have attributed to pressure ridges in lava flows. In a lacustrine context, they also resemble pressure ridges in desiccated evaporite deposits and salt-rimmed pools (now dry) similar in scale and morphology to spectacular, hundred meter-scale pool rims in alkaline Lake Natron, East African Rift. The eroded highland margin surface adjacent to these plains appears to be fairly smooth, even at 15 m/pixel. Isolated knob inliers are scattered from a few kilometers to several tens of "kilometers apart. Heights of the knobs have not been measured yet but, based on experience with similar features in the Pathfinder landing ellipse, are probably typically on the order of several tens of meters high and smaller, though some of the largest knobs in the region are probably up to a few hundred meters high. Two craters larger than a kilometer in diameter, with fluidized deposits, lie nearby the proposed landing site. Very high-resolution images from MOC should help to determine whether a landing site navigable by the Athena rover could be placed in this region. The space between knobs and craters is large enough to enable placement of a target landing ellipse between them but still provide access to one or more of them and to the margin of the Elysium plains material. Post-2001 Mars Surveyor Landing Site: Argyre Planitia (Proposed by T. J. Parker) Vital Statistics: (1) Latitude, Longitude: 55-56 S, 41-43 W. (2) Elevation (Viking): 1.0 km. (3) Viking Orbiter Image coverage: Excellent coverage by 40 m/pixel images (orbits 567B, 568B, and 569B). Excellent stereo coverage with large parallax angles over the entire landing site region, and much of central and southern Argyre. (4) Albedo: about .23-.24 (5) Block Abundance: No data (6) Fine-Component Thermal Inertia: No data The floors of both the Argyre and Hellas basins contain etched layered materials that are probably thick accumulations of channel or lacustrine sediments. The deposits in Hellas are much more eroded than those in Argyre, and Hellas lacks a channel outlet. Argyre is unique in that Uzboi Vallis flowed out of the basin, requiring overflow of a standing body of water within Argyre. This makes it the largest impact basin on Mars with channels both draining into it and flowing out from it. Hellas' channels may be catastrophic flood channels, whereas Argyre was fed by modest-scale valley networks, though the outlet at Uzboi Vallis was a catastrophic flood Highland craters and basins of this kind should be high-priority landing targets for missions intended to focus on the search for either prebiotic organic materials or even simple fossil microorganisms. Basins with internally-draining valley networks should be preferred over flood channels, as they could have provided the long-term influx of water favorable to the origin of life. (Catastrophic floods are not conducive to fossil preservation, due to their very short durations and high transportation energies). They also afford an opportunity to study the evolution of the planet's climate and volatiles during the period of time between the late Noachian and early Hesperian, when a drastic change from a proposed early warm, wet climate to one more closely resembling the modern environment is thought to have occurred. Large basin

Parker, T. J.↗

Performance Evaluation of the NASA/KSC Transmission System

NASA-KSC currently uses three bridged 100-Mbps FDDI segments as its backbone for data traffic. The FDDI Transmission System (FTXS) connects the KSC industrial area, KSC launch complex 39 area, and the Cape Canaveral Air Force Station. The report presents a performance modeling study of the FTXS and the proposed ATM Transmission System (ATXS). The focus of the study is on performance of MPEG video transmission on these networks. Commercial modeling tools - the CACI Predictor and Comnet tools - were used. In addition, custom software tools were developed to characterize conversation pairs in Sniffer trace (capture) files to use as input to these tools. A baseline study of both non-launch and launch day data traffic on the FTXS is presented. MPEG-1 and MPEG-2 video traffic was characterized and the shaping of it evaluated. It is shown that the characteristics of a video stream has a direct effect on its performance in a network. It is also shown that shaping of video streams is necessary to prevent overflow losses and resulting poor video quality. The developed models can be used to predict when the existing FTXS will 'run out of room' and for optimizing the parameters of ATM links used for transmission of MPEG video. Future work with these models can provide useful input and validation to set-top box projects within the Advanced Networks Development group in NASA-KSC Development Engineering.

Christensen, Kenneth J.↗

Towards Understanding the Fluid Dynamic Phenomenon of Interest to Rocket Base Heating: A Review

The significance of the base heating problem for rockets during ascent is due to the complex interaction between the rocket nozzle plumes and the external-flow which can change the flow characteristics in the base region dramatically. At lower altitudes the external-flow merges with the plume-flow, without the formation of a large separated flow region, and the cooler external-flow promotes convective cooling of the base wall. Under these conditions the majority of the base heating is due to radiative heating from the shock heated plume gases. At higher altitudes, however, the process of base heating is not so straightforward. The plume and the base flow expands dramatically and separated flow regions occur in the base area. Hot exhaust gases from the rocket nozzle will be entrained into the separated flow regions and produce a convective component to the base wall heating. Further, if the rocket exhaust-gas contains soot, the soot can increase the emission from the gas and dramatically increase the wall absorption coefficient for radiative heating if it is deposited on the walls . In addition, if the rocket exhaust gas is fuel rich, the fuel can bum in the separated flow regions and further increase the base heating. The base burning phenomenon, and the increased base heating caused by it at higher altitudes, have been observed for the Space Shuttle and Saturn Rocket. Under these conditions, the total heating is significantly higher than the heating without separated flow in the base region, and the increase in heating is directly attributable to the fluid dynamic complexity of the base region. Realistic simulation of the base heating requires that the calculated flow environment reproduce the fluid dynamic flow features accurately. Thus, it will be necessary to introduce into the CFD codes the capability for the flow to respond to the complex vehicle geometry, the effect of turbulence, the ability to accurately reproduce the plume shock/shear layer structures and capture such phenomenon as shock induced base separation and base-burning phenomenon. A survey of experimental, theoretical and computational work that details the fluid dynamics of the base flow environment will be presented in the proposed paper. CFD simulations of rocket base flows using standard CFD codes such as OVERFLOW or GASP will be explored to capture these phenomenon accurately. Merits and limitations of these codes for base flow environment predictions will be explored.

Venkatapathy, E.↗

Navier-Stokes Flowfield Simulation of Boeing 747-200 as Platform for SOFIA

Steady and unsteady viscous, three-dimensional flowfields are calculated using a thin layer approximation of Navier-Stokes equations in conjunction with Chimera overset grids. The finite-difference numerical scheme uses structured grids and a pentadiagonal flow solver called "OVERFLOW". The configuration of Boeing 747-200 has been chosen as one of configurations to be used as a platform for the SOFIA (Stratospheric Observatory For Infrared Astronomy). Initially, the steady flowfield of the full aircraft is calculated for the clean configuration (without a cavity to house telescope). This solution is then used to start the unsteady flowfield of a configuration containing cavity housing the observation telescope and its peripheral units. Analysis of unsteady flowfield in the cavity and its influence on the tail empennage, as well as the noise due to turbulence and optical quality of the flow are the main focus of this study. For the configuration considered here, the telescope housing cavity is located slightly downstream of the portwing. The entire flow-field is carefully constructed using 45 overset grids and consists of nearly 4 million grid points. All the computations axe done at one freestream flow condition of M(sub infinity) = 0.85, alpha = 2.5deg, and a Reynolds of Re = 1.85x10deg

Srinivasan, G.R.↗

Computational Modeling of the Ames 11-Ft Transonic Wind Tunnel in Conjunction with IofNEWT

Technical advances in Computational Fluid Dynamics have now made it possible to simulate complex three-dimensional internal flows about models of various size placed in a Transonic Wind Tunnel. TWT wall interference effects have been a source of error in predicting flight data from actual wind tunnel measured data. An advantage of such internal CFD calculations is to directly compare numerical results with the actual tunnel data for code assessment and tunnel flow analysis. A CFD capability has recently been devised for flow analysis of the NASA/Ames 11-Ft TWT facility. The primary objectives of this work are to provide a CFD tool to study the NASA/Ames 11-Ft TWT flow characteristics, to understand the slotted wall interference effects, and to validate CFD codes. A secondary objective is to integrate the internal flowfield calculations with the Pressure Sensitive Paint data, a surface pressure distribution capability in Ames' production wind tunnels. The effort has been part of the Ames IofNEWT, Integration of Numerical and Experimental Wind Tunnels project, which is aimed at providing further analytical tools for industrial application. We used the NASA/Ames OVERFLOW code to solve the thin-layer Navier-Stokes equations. Viscosity effects near the model are captured by Baldwin-Lomax or Baldwin-Barth turbulence models. The solver was modified to model the flow behavior in the vicinity of the tunnel longitudinal slotted walls. A suitable porous type wall boundary condition was coded to account for the cross-flow through the test section. Viscous flow equations were solved in generalized coordinates with a three-factor implicit central difference scheme in conjunction with the Chimera grid procedure. The internal flow field about the model and the tunnel walls were descretized by the Chimera overset grid system. This approach allows the application of efficient grid generation codes about individual components of the configuration; separate minor grids were developed to resolve the model and overset onto a main grid which discretizes the interior of the tunnel test section. Individual grid components axe not required to have mesh boundaries joined in any special way to each other or to the main tunnel grid. Programs have been developed to rotate the model about the tunnel pivot point and rotation axis, similar to that of the tunnel turntable mechanism for adjusting the pitch of the physical model in the test section.

Djomehri, M. Jahed↗

Algorithmic Enhancements for Unsteady Aerodynamics and Combustion Applications

Research in the FY01 focused on the analysis and development of enhanced algorithms for unsteady aerodynamics and chemically reacting flowfields. The research was performed in support of NASA Ames' efforts to improve the capabilities of the in-house computational fluid dynamics code, OVERFLOW. Specifically, the research was focused on the four areas: (1) investigation of stagnation region effects; (2) unsteady preconditioning dual-time procedures; (3) dissipation formulation for combustion; and (4) time-stepping methods for combustion.

Venkateswaran, Sankaran↗

The 1998 Outburst of XTE J1550-564: A Model Based on Multi-Wavelength Observations

The 1998 September outburst of the X-ray binary XTE J1550-564 was monitored at X-ray, optical and radio wavelengths. We divide the outburst sequence into five phases and discuss their multi-wavelength properties. The outburst starts with a hard X-ray spike, while the soft X-ray flux rises with a longer timescale. We suggest that the onset of the outburst is determined by an increased mass transfer rate from the companion star, but the outburst morphology is determined by the distribution of specific angular momentum in the accreting matter. The companion in XTE J1550-564 is likely an active magnetic star, with a surface field strong enough to influence the dynamics of mass transfer. We suggest that its magnetic field can create a magnetic bag capable of confining gas inside the Roche lobe of the primary. The impulsive rise in the hard X-rays is explained by the inflow of material with low angular momentum onto the black hole, on a free-fall timescale, when the magnetic confinement breaks down. At the same time, high angular momentum matter, transferred via ordinary Roche-lobe overflow, is responsible for the formation of a disk.

Wu, K.↗

CFD Process Automation Using Overset Grids

This talk summarizes three applications of the overset grid method for CFD using some level of automated grid generation, flow solution and post-processing. These applications are 2D high-lift airfoil analysis (INS2D code), turbomachinery applications (ROTOR2/3 codes), and subsonic transport wing/body configurations (OVERFLOW code). These examples provide a forum for discussing the advantages and disadvantages of overset gridding for use in an automated CFD process. The goals and benefits of the automation incorporated in each application will be described, as well as the shortcomings of the approaches.

Buning, Pieter G.↗

Evaluation of Aircraft Platforms for SOFIA by Computational Fluid Dynamics

The selection of an airborne platform for the Stratospheric Observatory for Infrared Astronomy (SOFIA) is based not only on economic cost, but technical criteria, as well. Technical issues include aircraft fatigue, resonant characteristics of the cavity-port shear layer, aircraft stability, the drag penalty of the open telescope bay, and telescope performance. Recently, two versions of the Boeing 747 aircraft, viz., the -SP and -200 configurations, were evaluated by computational fluid dynamics (CFD) for their suitability as SOFIA platforms. In each configuration the telescope was mounted behind the wings in an open bay with nearly circular aperture. The geometry of the cavity, cavity aperture, and telescope was identical in both platforms. The aperture was located on the port side of the aircraft and the elevation angle of the telescope, measured with respect to the vertical axis, was 500. The unsteady, viscous, three-dimensional, aerodynamic and acoustic flow fields in the vicinity of SOFIA were simulated by an implicit, finite-difference Navier-Stokes flow solver (OVERFLOW) on a Chimera, overset grid system. The computational domain was discretized by structured grids. Computations were performed at wind-tunnel and flight Reynolds numbers corresponding to one free-stream flow condition (M = 0.85, angle of attack alpha = 2.50, and sideslip angle beta = 0 degrees). The computational domains consisted of twenty-nine(29) overset grids in the wind-tunnel simulations and forty-five(45) grids in the simulations run at cruise flight conditions. The maximum number of grid points in the simulations was approximately 4 x 10(exp 6). Issues considered in the evaluation study included analysis of the unsteady flow field in the cavity, the influence of the cavity on the flow across empennage surfaces, the drag penalty caused by the open telescope bay, and the noise radiating from cavity surfaces and the cavity-port shear layer. Wind-tunnel data were also available to compare to the CFD results; the data permitted an assessment of CFD as a design tool for the SOFIA program.

Klotz, S. P.↗

Efficacy of Code Optimization on Cache-based Processors

The current common wisdom in the U.S. is that the powerful, cost-effective supercomputers of tomorrow will be based on commodity (RISC) micro-processors with cache memories. Already, most distributed systems in the world use such hardware as building blocks. This shift away from vector supercomputers and towards cache-based systems has brought about a change in programming paradigm, even when ignoring issues of parallelism. Vector machines require inner-loop independence and regular, non-pathological memory strides (usually this means: non-power-of-two strides) to allow efficient vectorization of array operations. Cache-based systems require spatial and temporal locality of data, so that data once read from main memory and stored in high-speed cache memory is used optimally before being written back to main memory. This means that the most cache-friendly array operations are those that feature zero or unit stride, so that each unit of data read from main memory (a cache line) contains information for the next iteration in the loop. Moreover, loops ought to be 'fat', meaning that as many operations as possible are performed on cache data-provided instruction caches do not overflow and enough registers are available. If unit stride is not possible, for example because of some data dependency, then care must be taken to avoid pathological strides, just ads on vector computers. For cache-based systems the issues are more complex, due to the effects of associativity and of non-unit block (cache line) size. But there is more to the story. Most modern micro-processors are superscalar, which means that they can issue several (arithmetic) instructions per clock cycle, provided that there are enough independent instructions in the loop body. This is another argument for providing fat loop bodies. With these restrictions, it appears fairly straightforward to produce code that will run efficiently on any cache-based system. It can be argued that although some of the important computational algorithms employed at NASA Ames require different programming styles on vector machines and cache-based machines, respectively, neither architecture class appeared to be favored by particular algorithms in principle. Practice tells us that the situation is more complicated. This report presents observations and some analysis of performance tuning for cache-based systems. We point out several counterintuitive results that serve as a cautionary reminder that memory accesses are not the only factors that determine performance, and that within the class of cache-based systems, significant differences exist.

VanderWijngaart, Rob F.↗