Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “data center efficiency”

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 1,495 records · Page 83
All sourcesNASA NTRSDOE OSTI
NASA NTRS1993

Pinatubo and Pre-Pinatubo Optical-Depth Spectra: Mauna Loa Measurements, Comparisons, Inferred Particle Size Distributions, Radiative Effects, and Relationship to Lidar Data

The Ames airborne tracking sunphotometer was operated at the National Oceanic and Atmospheric Administration (NOAA) Mauna Loa Observatory (MLO) in 1991 and 1992 along with the NOAA Climate Monitoring and Diagnostics Laboratory (CMDL) automated tracking sunphotometer and lidar. June 1991 measurements provided calibrations, optical-depth spectra, and intercomparisons under relatively clean conditions; later measurements provided spectra and comparisons for the Pinatubo cloud plus calibration checks. June 1991 results are similar to previous MLO springtime measurements, with midvisible particle optical depth tau(sub p)(lambda = 0.526 microns) at the near-background level of 0.012 +/- 0.006 and no significant wavelength dependence in the measured range (lambda = 0.38 to 1.06 microns). The arrival of the Pinatubo cloud in July 1991 increased midvisible particle optical depth by more than an order of magnitude and changed the spectral shape of tau(sub p)(lambda) to an approximate power law with an exponent of about -1.4. By early September 1991, the spectrum was broadly peaked near 0.5 microns, and by July 1992, it was peaked near 0.8 microns. Our optical-depth spectra include corrections for diffuse light which increase postvolcanic midvisible tau(sub p) values by 1 to 3% (i.e., 0.0015 to 0.0023). NOAA- and Ames Research Center (ARC)-measured spectra are in good agreement. Columnar size distributions inverted from the spectra show that the initial (July 1991) post-Pinatubo cloud was relatively rich in small particles (r less than 0.25 microns), which were progressively depleted in the August-September 1991 and July 1992 periods. Conversely, both of the later periods had more of the optically efficient medium-sized particles (0.25 less than r less than 1 micron) than did the fresh July 1991 cloud. These changes are consistent with particle growth by condensation and coagulation. The effective, or area-weighted, radius increased from 0.22 +/- 0.06 micron in July 1991 to 0.56 +/- 0.12 micron in August-September 1991 and to 0.86 +/- 0.29 micron in July 1992. Corresponding column mass values were 4.8 +/- 0.7, 9.1 +/- 2.7, and 5.5 +/- 2 micro g/sq cm, and corresponding column surface areas were 4.4 +/- 0.5, 2.9 +/- 0.2, and 1.1 +/- 0.1 sq micron/sq cm,. Photometer-inferred column backscatter values agree with those measured by the CMDL lidar on nearby nights. Combining lidar-measured backscatter profiles with photometer-derived backscatter-to-area ratios gives peak particle areas that could cause rapid heterogeneous loss of ozone, given sufficiently low particle acidity and suitable solar zenith angles (achieved at mid- to high latitudes). Top-of-troposphere radiative forcings for the September 1991 and July 1992 optical depths and size distributions over MLO are about -5 and -3 W 1/sq m, respectively (hence comparable in magnitude but opposite in sign to the radiative forcing caused by the increase in manmade greenhouse gases since the industrial revolution). Heating rates in Pinatubo layer over MLO are 0.55 +/- 0.13 and 0.41 +/- 0.14 K/d for September 1991 and July 1992, respectively.

Russell, P. B.↗
NASA NTRS1998

Mechanisms of cardiac hypertrophy in canine volume overload

This study tested whether the modest hypertrophy that develops in dogs in response to mitral regurgitation is due to a relatively small change in the rate of protein synthesis or, alternatively, is due to a decreased rate of protein degradation. After 3 mo of severe experimental mitral regurgitation, the left ventricular (LV) mass-to-body weight ratio increased by 23% compared with baseline values. This increase in LV mass occurred with a small, but not statistically significant, increase in the fractional rate of myosin heavy chain (MHC) synthesis (Ks), as measured using continuous infusion with [3H]leucine in dogs at 2 wk, 4 wk, and 3 mo after creation of severe mitral regurgitation. Translational efficiency was unaffected by mitral regurgitation as measured by the distribution of MHC mRNA in polysome gradients. Furthermore, there was no detectable increase in translational capacity as measured by either total RNA content or the rate of ribosome formation. These data indicate that translational mechanisms that accelerate the rate of cardiac protein synthesis are not responsive to the stimulus of mitral regurgitation. Most of the growth after mitral regurgitation was accounted for by a decrease in the fractional rate of protein degradation, calculated by subtracting fractional rates of protein accumulation at each time point from the corresponding Ks values. We conclude that 1) volume overload produced by severe mitral regurgitation does not trigger substantial increases in the rate of protein synthesis and 2) the modest increase in LV mass results primarily from a decrease in the rate of protein degradation.

Non-NASA Center↗
DOE OSTI2025

Development of Hydrogen Burner for FT4000® Aeroderivative Engine - Final Report

This report details an effort to develop a retrofittable fuel/air injector for the FT4000® aeroderivative gas turbine that enables use of hydrogen as a carbon-free fuel for efficient power generation. The FT4000 engine’s low-NOx combustor was developed by Pratt & Whitney and RTX Technology Research Center with core technology from the Pratt & Whitney PW4000 turbofan aircraft engine. The current FT4000 production engine operates on either natural gas or No. 2 fuel oil with water injection to achieve high thermal efficiency and low emissions. This engine is fielded by Mitsubishi Power Aero and delivers 70 MW of power with a simple-cycle efficiency of over 41% when operating with wet compression. The work reported here advances the technology readiness level of the FT4000 combustor for operation with hydrogen, starting with an experimental assessment of the current production hardware with increasing hydrogen content mixed with natural gas and ending with improved nozzle concepts for fully robust operation with 100% hydrogen. High-pressure single-sector combustor rig tests have been completed, demonstrating the ability for the dual fuel nozzle to operate an FT4000 combustor on 100% hydrogen with low nitrogen oxide (NOx) emissions. Metal temperature measurements and video images of the flame structure from zero to 100% hydrogen highlight opportunities to improve the fuel nozzle robustness for high hydrogen conditions. The design of new fuel/air mixer concepts to improve durability and operability with high hydrogen levels was also completed. A total of eleven new concepts were developed and analyzed, ranging from modifications to the bill-of-materials nozzle to fully clean sheet designs. The concepts were evaluated with non-reacting and reacting flow evaluations to assess the performance of the new hardware designs. Non-reacting tests included Phase Doppler Particle Analysis (PDPA) for droplet size and velocity, mechanical patternation for liquid water flux, and planar laser induced fluorescence (PLIF) with acetone-seeding for gaseous fuel/air mixing characterization. Five scaled candidate fuel nozzle designs, in addition to a scaled bill-of-materials nozzle, were then successfully evaluated in an atmospheric pressure burner rig. The nozzles were evaluated for performance with natural gas, hydrogen/natural gas blends, and pure hydrogen. For 100% hydrogen, the nozzles were evaluated with and without water injection. Optical and infrared imaging of the flame and fuel nozzle was captured. NOx emissions were sampled from fixed emissions probes. The results show measurable differences between the various designs, and the data was used to down-select the two most promising designs to advance to future full pressure rig testing. Results from this study have cleared the current production FT4000 engines with dual fuel nozzles to operate at baseload power on blends of hydrogen mixed with natural gas and water. Two promising nozzle designs have been developed to enable fully robust operation with up to 100% hydrogen. These nozzles require validation at full baseload operating conditions before they can be introduced to the field.

03 NATURAL GAS↗
NASA NTRS1994

The SAX Italian scientific satellite. The on-board implemented automation as a support to the ground control capability

This paper presents the capabilities implemented in the SAX system for an efficient operations management during its in-flight mission. SAX is an Italian scientific satellite for x-ray astronomy whose major mission objectives impose quite tight constraints on the implementation of both the space and ground segment. The most relevant mission characteristics require an operative lifetime of two years, performing scientific observations both in contact and in noncontact periods, with a low equatorial orbit supported by one ground station, so that only a few minutes of communications are available each orbit. This operational scenario determines the need to have a satellite capable of performing the scheduled mission automatically and reacting autonomously to contingency situations. The implementation approach of the on-board operations management, through which the necessary automation and autonomy are achieved, follows a hierarchical structure. This has been achieved adopting a distributed avionic architecture. Nine different on-board computers, in fact, constitute the on-board data management system. Each of them performs the local control and monitors its own functions while the system level control is performed at a higher level by the data handling applications software. The SAX on-board architecture provides the ground operators with different options of intervention by three classes of telecommands. The management of the scientific operations will be scheduled by the operation control center via dedicated operating plans. The SAX satellite flight mode is presently being integrated at Alenia Spazio premises in Turin for a launch scheduled for the end of 1995. Once in orbit, the SAX satellite will be subject to intensive check-out activities in order to verify the required mission performances. An overview of the envisaged procedure and of the necessary on-ground activities is therefore depicted as well.

Martelli, Andrea↗
DOE OSTI2021

A Massively Parallel Implementation of the CCSD(T) Method Using the Resolution-of-the-Identity Approximation and a Hybrid Distributed/Shared Memory Parallelization Model

In this work, a parallel algorithm is described for the coupled-cluster singles and doubles method augmented with a perturbative correction for triple excitations [CCSD(T)] using the resolution-of-the-identity (RI) approximation for two-electron repulsion integrals (ERIs). The algorithm bypasses the storage of four-center ERIs by adopting an integral-direct strategy. The CCSD amplitude equations are given in a compact quasi-linear form by factorizing them in terms of amplitude-dressed three-center intermediates. A hybrid MPI/OpenMP parallelization scheme is employed, which uses the OpenMP-based shared memory model for intranode parallelization and the MPI-based distributed memory model for internode parallelization. Parallel efficiency has been optimized for all terms in the CCSD amplitude equations. Two different algorithms have been implemented for the rate-limiting terms in the CCSD amplitude equations that entail and -scaling computational costs, where N O and N V denote the number of correlated occupied and virtual orbitals, respectively. One of the algorithms assembles the four-center ERIs requiring N V 4 and N O 2 N V 2 -scaling memory costs in a distributed manner on a number of MPI ranks, while the other algorithm completely bypasses the assembling of quartic memory-scaling ERIs and thus largely reduces the memory demand. It is demonstrated that the former memory-expensive algorithm is faster on a few hundred cores, while the latter memory-economic algorithm shows a better strong scaling in the limit of a few thousand cores. The program is shown to exhibit a near-linear scaling, in particular for the compute-intensive triples correction step, on up to 8000 cores. The performance of the program is demonstrated via calculations involving molecules with 24–51 atoms and up to 1624 atomic basis functions. As the first application, the complete basis set (CBS) limit for the interaction energy of the π-stacked uracil dimer from the S66 data set has been investigated. This work reports the first calculation of the interaction energy at the CCSD(T)/aug-cc-pVQZ level without local orbital approximation. The CBS limit for the CCSD correlation contribution to the interaction energy was found to be -8.01 kcal/mol, which agrees very well with the value -7.99 kcal/mol reported by Schmitz, Hättig, and Tew [ Phys. Chem. Chem. Phys. 2014 , 16 , 22167-22178]. The CBS limit for the total interaction energy was estimated to be -9.64 kcal/mol.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2022

22” ADP Fan Rig Liners Design Report

A liner design study was conducted as part of a cooperative effort between five (5) government/industry teams that together seek to demonstrate the technology for designing and manufacturing acoustic liners that are twenty five percent (25%) more efficient than 1992 technology liners. The study emphasized teaming collaboration. The improved liners were designed by Pratt & Whitney and Boeing Airplane Company and built by Rohr Inc, and will be tested in the NASA/P& W 22-inch ADP fan rig at NASA Lewis Research Center's 9' x 15' wind tunnel. Design guidelines and decisions were made collectively during monthly design review telecons and at the formal final design review. The tools that were used to design the new improved liners were not new, but the process that was developed as part of this study that led into the evaluation and selection of the final and best liner designs is new. Until now, the liner design process as practiced by different industry teams varied greatly. Some procedures were based on empirical liner attenuation databases which could not adequately account for engine-to-engine hardwall fan noise spectral differences, while others were entirely theoretical. And regardless of which method one chose, the major difficulty was still the lack of understanding and knowledge of the actual hardwall fan source noise modal structure and farfield SPL spectra. The NASA-led effort to attempt actual measurements of the fan source noise modal structure by means of the rotating microphone array is expected to contribute significantly to the understanding of the nature of the fan tone noise modes, but the application to broadband noise is still a long way off. The new design process included consideration of the measured fan tone modes, but for the majority of the spectra a separate, systematic process was used. It is a common practice in the engine/nacelle industry that acoustic liners for new products are designed before actual measured hardwall engine far-field noise spectra are available. Target noise spectra are normally derived from existing engine noise databases with some adjustments to absolute SPLs and frequencies. This practice has been acceptable as long as the new engine was a derivative of the base engine. However, in the case of the ADP, the transition from a current engine base is too - great, and the adjustments could not adequately account for the quantum changes in SPL spectral differences. Examples were the 1992 single degree of freedom (SDOF) inlet and aft liners (designed by P&W) that would be used as the baseline liners against which the new improved liners' efficien¬cies would -be measured. As will be shown, these baseline liners have been found to be deeper than the desire optimum depths. The new liner design process begins with the selection of the target hard wall fan noise spectra. These spectra were obtained by scaling up (5.91 scale factor) the measured hardwall fan spectral data from the 22” ADP fan rig. Next, the modal energy contents for each 1/3-octave band center frequency of these hardwall fan noise spectra are estimated. (Within each 1/3-octave band center frequency, the model energy distribution approximation is for both tone and broadband noise). For the inlet noise, P&W uses a derivative of the NASA Lewis (Ed Rice's) method of classifying; and grouping propagating modes by their cutoff ratios. The next step is to assign energy level to each group of modes having the same cutoff ratio. In Ed Rice's model, the modes were grouped according to ten (10) cutoff ratio intervals with center values located at 1.026, 1.085, 1.155, 1.24, 1.35, 1.49, 1.69, 2.0, and 4.47 (with equal number of modes in each interval). The modes were assumed to have equal energy. P&W’s model expands the cutoff ratio-mode grouping into two hundred (200) smaller cutoff ratio intervals with center values located f1".'m 1.003 to 11.5 in 199 increasing incremental intervals. Next, P&W's model uses a "2-parameter" normal distribution as a template to assign energy levels to these 200 pre-determined cutoff ratio values (for each 1/3-octave band center frequency). In the past, P&W had conducted an extensive study to determine what "2-parameter" values are appropriate for fullscale inlet liners, and had developed a set of twenty-four (24) "2-pararneter" values (i.e. one for each 1/3-octave frequency band) that when used in Ed Rice's Inlet Attenuation Prediction Method produced predicted liner attenuations that closely matched measured liner attenuations from several P&W's engines. In the absence of actual measured tone and broadband modal data from the 22" ADP rig, the process will use P&W's proprietary set of "2-parameter" values for this liner design study. For the aft noise, Boeing uses a modal energy approximation that the "transport energy" of each propagating mode is equal. This approximation is almost the same as the equal energy per mode approximation, except for modes that are near cutoff. Boeing's model forces these modes to have lower energy levels. Both P&W and Boeing agreed that the "transport energy" approximation should work well for the ADP aft fan noise which appears to be dominated by broadband noise. The design process then proceeds to calculate the optimum liner impedances for each frequency in both the inlet and the aft. These optimum impedances represent the target impedances that the designed liners should have. Liners with impedances matching the optimum impedances at all frequencies are "ideal" liners. These ideal liners are theoretically the best liners. Unfortunately, it has been showed that it is impossible to design and build such ideal liners. The next best liners are ones that have impedances matching the optimum impedances over some frequency range (not all frequencies as for the ideal). This is accomplished by the use of "frequency weightings". Several of P&W's and Boeing's existing computer decks were used for optimizing and matching the designed liner impedances to the target optimum values (with the various frequency weightings specified). The optimization produces liner candidates with predicted liner impedances and descriptions of their physical liner characteristics. These candidate liner impedances are then used to predict their spectral attenuation characteristics which are then used together with the target hardwall fan noise spectra to determine the resulting treated noise spectra and PNLT values. Further optimization around the selected candidate designs yield final designs that are best in PNLT attenuations. Use of the optimization decks allowed a large number of liner candidates to be screened in a relatively short period of time. This design process is systematic and is efficient. The new inlet SDOF liner design obtained from the improved process was predicted to be 34% more effective (per unit area) than the 1992 baseline inlet liner. This inlet liner design is a 112 rayl wovenwiremesh facesheet over a 0.312-inch deep honeycomb core. The new aft SDOF liner design was predicted to be 52% more efficient than the aft baseline liner. The new aft SDOF liner is "segmented" with a shallower liner on the core cowl (inner duct wall) and a deeper liner on the fan cowl (outer duct wall). The shallower liner is a 70.6 rayl woven-wiremesh facesheet over a 0.141-inch deep honeycomb core. The deeper liner is a 68 rayl woven-wiremesh facesheet over a 0.309-inch deep honeycomb core. All liner dimensions are for the 22-inch model-scale ADP fan rig liners. The selected advanced liners are "segmented" double-layer (DDOF) liners for the inlet and aft locations. Also, for the inlet, a bulk liner with ceramic foam for wider broadband noise absorption was also selected. Triple-layer liner designs were not considered since the model scaled liners ( 1/5. 91) were dimensionally too small to be built correctly, and irrin earlier concept study, Boeing found a triple-layer to have only very small benefits over a double-layer. The inlet DDOF liner was predicted to be 83% more effective than the baseline. The inlet DDOF design is a 78 rayl facesheet over a 0.080 top cavity depth, a 68 rayl septum and a 0.227-inch bottom cavity depth. The inlet bulk liner is a 60 rayl facesheet over a 0.33-inch deep honeycomb filled with high temperature (HTP) ceramic foam with a density of 4.8 lb/cu.ft and a flow resistivity of 167 rayl/cm. The inlet bulk liner was predicted to be 83% more effective than the baseline liner. The aft DDOF liners are segmented with a shallower liner on the core cowl and a deeper liner on the fan cowl. The shallow DDOF liner is a 49.8 rayl facesheet over a 0.093-inch top cavity depth, a septum of 88.2 rayls over a 0.181-inch bottom cavity depth. The deep DDOF liner is a 12.9 rayl facesheet over a 0.140-inch top cavity depth, a septum of 53.1 ray ls over a 0.258-inch botom cavity depth. The segmented aft DDOF liners were predicted to be 86% more efficient than the baseline liner.

turbofan acoustic treatment↗
NASA NTRSDate not supplied

Forward and Inverse Models for Satellite Remote Sensors using Principal Component Analysis

Satellite remote sensors such as AIRS on Aqua, CrIS on S-NPP, NOAA20 and JPSS-2, IASI on Metop A, B, and C make millions of observations each day with thousands of spectral channels for each observation; this poses challenges for efficiently inversion of the inherently large dataset as needed to retrieve atmospheric and surface properties. This presentation will illustrate the use of Principal Component Analysis (PCA) to speed up radiative transfer forward model calculations and to stabilize the inversion algorithms. A Principal Component-based radiative transfer model (PCRTM) developed at NASA Langley Research Center can simulate top of atmosphere (TOA) radiance or reflectance spectra from 50 cm-1 to 50000 cm-1 (200 m to 0.20 m quickly and accurately. PCRTM demonstrated very high accuracy relative to reference line-by-line radiative transfer models and it saves orders of magnitude computational time. Examples of the PCRTM model developed for hyperspectral sensors such as AIRS, CrIS, IASI, NAST-I, SHIS, CPF, TEMPO, SBG, OMI, and SCIAMACHY will be presented. In addition to using the PCRTM as forward model, the NASA Langley developed inversion algorithm also uses PCA to compress the state vector into a compressed dimension to speed up and stabilize the inversion process. Examples of retrieved atmospheric temperature, water vapor, CO2, CO, CH4, N2O, and O3 profiles, cloud properties (optical depth, size, phase, and height), and surface properties (surface emissivity spectra and skin temperatures) will be presented. This algorithm is being transitioned to the NASA Sounder SIPS and NASA's Goddard Earth Sciences Data and Information Services Center (GES DISC).

forward model↗
NASA NTRS2003

Neural Network and Regression Methods Demonstrated in the Design Optimization of a Subsonic Aircraft

The neural network and regression methods of NASA Glenn Research Center s COMETBOARDS design optimization testbed were used to generate approximate analysis and design models for a subsonic aircraft operating at Mach 0.85 cruise speed. The analytical model is defined by nine design variables: wing aspect ratio, engine thrust, wing area, sweep angle, chord-thickness ratio, turbine temperature, pressure ratio, bypass ratio, fan pressure; and eight response parameters: weight, landing velocity, takeoff and landing field lengths, approach thrust, overall efficiency, and compressor pressure and temperature. The variables were adjusted to optimally balance the engines to the airframe. The solution strategy included a sensitivity model and the soft analysis model. Researchers generated the sensitivity model by training the approximators to predict an optimum design. The trained neural network predicted all response variables, within 5-percent error. This was reduced to 1 percent by the regression method. The soft analysis model was developed to replace aircraft analysis as the reanalyzer in design optimization. Soft models have been generated for a neural network method, a regression method, and a hybrid method obtained by combining the approximators. The performance of the models is graphed for aircraft weight versus thrust as well as for wing area and turbine temperature. The regression method followed the analytical solution with little error. The neural network exhibited 5-percent maximum error over all parameters. Performance of the hybrid method was intermediate in comparison to the individual approximators. Error in the response variable is smaller than that shown in the figure because of a distortion scale factor. The overall performance of the approximators was considered to be satisfactory because aircraft analysis with NASA Langley Research Center s FLOPS (Flight Optimization System) code is a synthesis of diverse disciplines: weight estimation, aerodynamic analysis, engine cycle analysis, propulsion data interpolation, mission performance, airfield length for landing and takeoff, noise footprint, and others.

Hopkins, Dale A.↗
NASA NTRS2006

An Operational Configuration of the ARPS Data Analysis System to Initialize WRF in the NM'S Environmental Modeling System

The Weather Research and Forecasting (WRF) model is the next generation community mesoscale model designed to enhance collaboration between the research and operational sectors. The NM'S as a whole has begun a transition toward WRF as the mesoscale model of choice to use as a tool in making local forecasts. Currently, both the National Weather Service in Melbourne, FL (NWS MLB) and the Spaceflight Meteorology Group (SMG) are running the Advanced Regional Prediction System (AIRPS) Data Analysis System (ADAS) every 15 minutes over the Florida peninsula to produce high-resolution diagnostics supporting their daily operations. In addition, the NWS MLB and SMG have used ADAS to provide initial conditions for short-range forecasts from the ARPS numerical weather prediction (NWP) model. Both NM'S MLB and SMG have derived great benefit from the maturity of ADAS, and would like to use ADAS for providing initial conditions to WRF. In order to assist in this WRF transition effort, the Applied Meteorology Unit (AMU) was tasked to configure and implement an operational version of WRF that uses output from ADAS for the model initial conditions. Both agencies asked the AMU to develop a framework that allows the ADAS initial conditions to be incorporated into the WRF Environmental Modeling System (EMS) software. Developed by the NM'S Science Operations Officer (S00) Science and Training Resource Center (STRC), the EMS is a complete, full physics, NWP package that incorporates dynamical cores from both the National Center for Atmospheric Research's Advanced Research WRF (ARW) and the National Centers for Environmental Prediction's Non-Hydrostatic Mesoscale Model (NMM) into a single end-to-end forecasting system. The EMS performs nearly all pre- and postprocessing and can be run automatically to obtain external grid data for WRF boundary conditions, run the model, and convert the data into a format that can be readily viewed within the Advanced Weather Interactive Processing System. The EMS has also incorporated the WRF Standard Initialization (SI) graphical user interface (GUT), which allows the user to set up the domain, dynamical core, resolution, etc., with ease. In addition to the SI GUT, the EMS contains a number of configuration files with extensive documentation to help the user select the appropriate input parameters for model physics schemes, integration timesteps, etc. Therefore, because of its streamlined capability, it is quite advantageous to configure ADAS to provide initial condition data to the EMS software. One of the biggest potential benefits of configuring ADAS for ingest into the EMS is that the analyses could be used to initialize either the ARW or NMM. Currently, the ARPS/ADAS software has a conversion routine only for the ARW dynamical core. However, since the NIvIM runs about 2.5 times faster than the ARW, it is quite advantageous to be able to run an ADAS/NMM configuration operationally due to the increased efficiency.

Case, Jonathan↗
NASA NTRS2000

Operational Considerations when Designing New Ground Systems

The Hubble Space Telescope (HST) launched in April of 1991 with a nominal 15-year old mission. Since then, the HST mission life has been extended to 2010. As is true for all NASA missions, HST is being asked to decrease its operational costs for the remainder of its mission life. Various techniques are being incorporated for cost reductions, with one of the core means being the design of a new and more efficient ground system for HST operations. This new ground system, "Vision 2000", will reduce operational and maintenance costs and also provide the HST Project with added flexibility to react to future changes. Vision 2000 began supporting HST Operations in January of 1999 and will support the mission for the remainder of the mission life. Upgrading a satellite's ground system is a popular approach for reducing costs, but it is also inherently risky. Validating a new ground system can be a severe distraction to a flight team while operating a satellite. Mission data collection and health and safety requirements are rarely, if ever, relaxed during this validation period, forcing flight teams to undertake an additional task while operating the satellite. Additionally, flight teams must usually undergo extensive training to effectively utilize the new system. Once again, this training usually occurs as an additional task, in addition to the nominal satellite operations. While operating the spacecraft, the Flight Team typically assists in the design, validation, and verification of a new ground system. This is a distraction and strain on the Flight Team, but the benefit of using the Flight Team in all phases of ground system development far outweigh the negative aspects. Finally, above the cost of the new system, the integration into the facility with the current control center system are resources and costs not normally taken into account in the design phase of the new system. In addition to the standard issues faced by a Project when upgrading its ground system, the HST Project also must continue to support Space Shuttle servicing missions, which occur approximately every two to three years. This paper will address many of the issues common to all missions when re-hosting ground systems, and those faced by the HST Project in particular.

Walyus, Keith↗
DOE OSTI2020

Network traffic control for multi-homed end-hosts via SDN

Software-defined networking (SDN) is an emerging technology of efficiently controlling and managing computer networks, such as in data centres, wide-area networks, as well as in ubiquitous communication. In this study, the authors explore the idea of embedding the SDN components, represented by SDN controller and virtual switch, in end-hosts to improve network performance. In particular, the authors consider load balancing across multiple network interfaces on end-hosts with different link capacity scenarios. The authors have explored and implemented different SDN-based load-balancing approaches based on OpenFlow software switches, and have demonstrated the feasibility and the potential of this approach. The proposed system has been evaluated with MultiPath transmission control protocol (MPTCP). Furthermore, the proposed results demonstrated the potential of applying the SDN concepts on multi-homed devices resulting in an increase in achieved throughput of 55% compared to the legacy single network approach and 10% compared to the MPTCP.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Evaluating Modern GPU Interconnect: PCIe, NVLink, NV-SLI, NVSwitch and GPUDirect

High performance multi-GPU computing becomes an inevitable trend due to the ever-increasing demand on computation capability in emerging domains such as deep learning, big data and planet-scale simulations. However, the lack of deep understanding on how modern GPUs can be connected and the real impact of state-of-the-art interconnect technology on multi-GPU application performance become a hurdle. In this paper, we fill the gap by conducting a thorough evaluation on five latest types of modern GPU interconnects: PCIe, NVLink-V1, NVLink-V2, NVLink-SLI and NVSwitch, from six high-end servers and HPC platforms: NVIDIA P100-DGX-1, V100-DGX-1, DGX-2, OLCF’s SummitDev and Summit supercomputers, as well as an SLI-linked system with two NVIDIA Turing RTX-2080 GPUs. Based on the empirical evaluation, we have observed four new types of GPU communication network NUMA effects: three are triggered by NVLink’s topology, connectivity and routing, while one is caused by PCIe chipset design issue. These observations indicate that, for a multi-GPU application running in a multi-GPU node, choosing the right GPU combination can impose considerable impact on GPU communication efficiency, as well as an application’s overall performance. Our evaluation can be leveraged in building practical multi-GPU performance models, which are vital for GPU task allocation, scheduling and migration in a shared environment (e.g., AI cloud and HPC centers), as well as communication-oriented performance tuning.

Performance Evaluation, GPU, Interconnect, NUMA, P↗
DOE OSTI2023

IceNet for FireBox - A Berkeley Warehouse-Scale Computer

Berkeley’s FireBox is a next-generation warehouse-scale computer (WSC) that utilizes the energy-efficiency and bandwidth density of integrated silicon-photonic interconnects to enable a new high-bandwidth and low-latency network fabric connecting thousands of compute nodes to petabytes of DRAM and Flash storage. The high bandwidth, low latency and high connectivity of FireBox’s WSC network fabric (IceNet) will enable dramatic improvements in the overall system energy efficiency enabling fine-grain power control on system resources (processors, links and memory/storage components). IceNet is a special 3-stage photonic Clos network architected to achieve ultra-low-latency connectivity between processor nodes and memory, drastically cutting down on the energy wasted in resource idling (processors and memory stalled due to pending network requests). This is achieved by integration of the first and last switch stages into processor/memory hub clients and by heavy over-provisioning of the high-radix middle switches (FlareSwitches). A key hardware component developed in this program is an active laser power management photonic integrated circuits called LightSpark. It interacts with the FlareSwitch and provides laser power to a subset of occupied switch ports, increasing the utilization of laser light in the photonic network by an order of magnitude. In addition to guiding the laser power where it is needed, the laser-power management module enables both wavelength and laser redundancy, significantly increasing the robustness of the system. The goal of the IceNet fabric is to enable communication between 1000s of processor nodes and PBs of memory/storage with <100ns latency, <10pJ/b wall-plug energy cost at multiple Pb/s of available connectivity bandwidth. These metrics represent two-orders of magnitude improvement with respect to the status of current data-center technology.

42 ENGINEERING↗
DOE OSTI2026

Uncertainty-Aware Machine Learning for Small-Angle X-ray Scattering Analysis in Autonomous Experimentation

Small-angle X-ray scattering (SAXS) is a powerful high-throughput characterization tool for probing nanoscale structure in native sample environments, providing real-time morphological information such as nanoparticle size and shape during synthesis. However, automated SAXS data analysis for extracting meaningful structural parameters is non-trivial and remains a bottleneck in closed-loop experimentation towards autonomous materials discovery, which demands fast, reliable, and uncertainty-aware data analysis. Here, we develop a machine-learning approach for automated SAXS analysis tailored to closed-loop nanoparticle synthesis. A Random Forest (RF) regression model is trained on 100,000 synthetic SAXS curves generated from polydisperse spherical nanoparticles with realistic background contributions. Using normalized one-dimensional SAXS intensity profiles as input, the RF model directly predicts nanoparticle radius, size polydispersity, and background parameters, while the ensemble standard deviation across trees provides built-in uncertainty quantification (UQ). On synthetic data, we show that combining fit-quality metrics (R 2 , MAE) with thresholds on prediction uncertainty reliably identifies accurate parameter estimates without access to ground truth. We then apply the trained model to 365 experimental SAXS profiles of citrate-reduced gold nanoparticles synthesized using an automated droplet-flow microreactor with in situ SAXS at a synchrotron beamline, classifying the results into high- and low-confidence subsets based on UQ metrics. Finally, we integrate RF-based SAXS analysis into a simulated closed-loop optimization campaign using Gaussian process Bayesian optimization to minimize nanoparticle polydispersity, benchmarking against conventional automated Levenberg–Marquardt fitting. The RF-guided campaign exhibits substantially faster convergence and lower relative opportunity cost (∼0.07 vs ∼0.3), demonstrating that uncertainty-aware machine-learning SAXS analysis significantly enhances the efficiency and robustness of autonomous nanomaterials synthesis workflows.

Bayesian optimization↗
DOE OSTI2020

Demonstration of μCHP in Light Commercial Hot Water Applications

Internal combustion engine driven combined heat and power (CHP) systems produce power while the waste heat is recovered and used for another purpose. For µCHP systems (less than 50kWe output) overall efficiencies can reach over 90% for products currently on the market throughout the world. On a fuel basis this offers an advantage over using grid-based power and another fuel for heat. Additionally, the cost of natural gas on an energy unit basis is significantly less expensive than electric power. This difference can lead to substantial savings for the end user of a CHP system. In some parts of the world, particularly Germany and Japan, µCHP has become a common part of the heat and distributed generation equipment mix. Market analysis conducted in 2012 showed that in North America there is potential of a 5,300 units per year market for µCHP, resulting in an annual energy savings of approximately 13.5 TBtu and more than $84 million in annual customer savings. However, µCHP still has not seen much market penetration in North America. This study helps to identify some of the reasons why, address roadblocks that still need to be overcome and provide experience to established best practices. Nine commercial sites were initially identified (fitness club, surgical center, senior community, resort and spa, commercial laundry, two multi-family housing, restaurant and manufacturing facility). Systems from two different manufacturers were installed at five sites total, but due to unforeseen circumstances, only three of those sites were finally commissioned: test data were collected by Oak Ridge National Laboratory. While not all systems were installed as intended, the project resulted in the following key findings: Micro combined heat and power is an unfamiliar technology for building owners and plumbing and electrical contractors that serve the buildings most suited for µCHP. Manufacturers will need to be aggressive in providing training and providing broad support to successfully bring the technology to market.; The proper integration of µCHP into other building systems is the most critical part of a successful installation. Manufacturers will need to be able to provide the expertise to train, advise and support the mechanical contractors during the design and installation of the heat integration system.; Regulations governing µCHP varies across the country. There are a few UL standards (UL 2200 and UL 1741) that apply nationally. Emissions are governed by EPA except in California where the California Air Resources Board (CARB) has established their own standard. Requirements that dictate installation and interconnection are dependent on local building codes and local utility processes and requirements. Some out of date building codes do not offer clear guidance for these requirements, leading to confusion. This makes developing training and determining broadly applicable best practices difficult.; Installation costs of µCHP are a key factor in making an installation successful. It is critical that these costs and all national, state and local regulations be understood and accounted for by all parties involved in the installation before a project is undertaken.; A robust customer service organization is important to meet the needs of the customers. These systems require annual maintenance and service and local capability is needed to keep costs inline.

03 NATURAL GAS↗
NASA NTRS1999

Parallel Grid Manipulations in Earth Science Calculations

The National Aeronautics and Space Administration (NASA) Data Assimilation Office (DAO) at the Goddard Space Flight Center is moving its data assimilation system to massively parallel computing platforms. This parallel implementation of GEOS DAS will be used in the DAO's normal activities, which include reanalysis of data, and operational support for flight missions. Key components of GEOS DAS, including the gridpoint-based general circulation model and a data analysis system, are currently being parallelized. The parallelization of GEOS DAS is also one of the HPCC Grand Challenge Projects. The GEOS-DAS software employs several distinct grids. Some examples are: an observation grid- an unstructured grid of points at which observed or measured physical quantities from instruments or satellites are associated- a highly-structured latitude-longitude grid of points spanning the earth at given latitude-longitude coordinates at which prognostic quantities are determined, and a computational lat-lon grid in which the pole has been moved to a different location to avoid computational instabilities. Each of these grids has a different structure and number of constituent points. In spite of that, there are numerous interactions between the grids, e.g., values on one grid must be interpolated to another, or, in other cases, grids need to be redistributed on the underlying parallel platform. The DAO has designed a parallel integrated library for grid manipulations (PILGRIM) to support the needed grid interactions with maximum efficiency. It offers a flexible interface to generate new grids, define transformations between grids and apply them. Basic communication is currently MPI, however the interfaces defined here could conceivably be implemented with other message-passing libraries, e.g., Cray SHMEM, or with shared-memory constructs. The library is written in Fortran 90. First performance results indicate that even difficult problems, such as above-mentioned pole rotation- a sparse interpolation with little data locality between the physical lat-lon grid and a pole rotated computational grid- can be solved efficiently and at the GFlop/s rates needed to solve tomorrow's high resolution earth science models. In the subsequent presentation we will discuss the design and implementation of PILGRIM as well as a number of the problems it is required to solve. Some conclusions will be drawn about the potential performance of the overall earth science models on the supercomputer platforms foreseen for these problems.

Sawyer, W.↗
NASA NTRS2005

Exhumation of Basement-cored Uplifts: Example of the Kyrgyz Range Quantified with Apatite Fission-track Thermochronology

The Kyrgyz Range, the northernmost portion of the Kyrgyzstan Tien Shan, displays topographic evidence for lateral propagation of surface uplift and exhumation. The highest and most deeply dissected segment lies in the center of the range. To the east, topography and relief decrease, and preserved remnants of a Cretaceous regional erosion surface imply minimal amounts of bedrock exhumation. The timing of exhumation of range segments defines the lateral propagation rate of the range-bounding reverse fault and quantifies the time and erosion depth needed to transform a mountain range from a juvenile to a mature morphology. New apatite fission-track (AFT) data from three transects from the eastern Kyrgyz Range, combined with published AFT data, demonstrate that the range has propagated over 110 km eastwards over the last 7-11 Myr. Based on the thermal and topographic evolutionary history, we present a model for a time-varying exhumation rate driven by rock uplift and changes in erodability and the time scale of geomorphic adjustment to surface uplift. Easily eroded, Cenozoic sedimentary rocks overlying resistant basement control early, rapid exhumation and slow surface upliftrates. As increasing amounts of resistant basement are exposed, exhumation rates decrease while surface uplift rates are sustained or increase, thereby growing topography. As the range becomes high enough to cause ice accumulation and develop steep river valleys, fluvial and glacial erosion become more powerful and exhumation rates once again increase. Independently determined range-noma1 shortening rates have also varied over time, suggesting a feedback between erosional efficiency and shortening rate.

Sobel, Edward R.↗
NASA NTRS2016

[EC5-Space Suit Assembly Team- Internship]

There were three main projects in this internship. The first pertained to the Bearing Dust Cycle Test, in particular automating the test to allow for easier administration. The second concerned modifying the communication system setup in the Z2 suit, where speakers and mics were adjusted to allow for more space in the helmet. And finally, the last project concerned the tensile strength testing of fabrics deemed as candidates for space suit materials and desired to be sent off for radiation testing. The major duties here are split up between the major projects detailed above. For the Bearing Dust Cycle Test, the first objective was to find a way to automate administration of the test, as the previous version was long and tedious to perform. In order to do this, it was necessary to introduce additional electronics and perform programming to control the automation. Once this was done, it would be necessary to update documents concerning the test setup, procedure, and potential hazards. Finally, I was tasked with running tests using the new system to confirm system performance. For the Z2 communication system modifications, it was necessary to investigate alternative speakers and microphones which may have better performance than those currently used in the suit. Further, new speaker and microphone positions needed to be identified to keep them out of the way of the suit user. Once this was done, appropriate hardware (such as speaker or microphone cases and holders) could be prototyped and fabricated. For the suit material strength testing, the first task was to gather and document various test fabrics to identify the best suit material candidates. Then, it was needed to prepare samples for testing to establish baseline measurements and specify a testing procedure. Once the data was fully collected, additional test samples would be prepared and sent off-site to undergo irradiation before being tested again to observe changes in strength performance. For the Bearing Dust Cycle Test, automation was achieved through use of a servo motor and code written in LabVIEW. With this a small electrical servo controller was constructed and added to the system. For the Z2 communication modifications speaker cases were developed and printed, and new speakers and mics were selected. This allowed us to move the speakers and mics to locations to remain out of the suit users way. For the suit material strength testing, five material candidates were identified and test samples were created. These samples underwent testing, and baseline test results were gathered, though these results are currently being investigated for accuracy. The main process efficiency developed during the course of this internship comes from automation of the Bearing Dust Cycle Test. In particular, many hours of human involvement and precise operation are replaced with a simple motor setup. Thus it is no longer required to man the test, saving valuable employee time. This internship has confirmed a few things for me, namely that I both want to work as an engineer for an aerospace firm and that in particular I want to work for the Johnson Space Center. I am also confirmed in my desire to work with electronics, though I was surprised to enjoy prototyping 3D CAD design as much as I did. Therefore, I will make an effort to build my skills in this area so that I can continue to design mechanical models. In fact, I found the process of hands-on prototyping to be perhaps the most fun aspect of my time working here. This internship has also furthered my excitement for continual education, and I will hopefully be pursuing a masters in my field in the near future.

Maicke, Andrew↗