Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “LOGS”

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 559 records · Page 31
All sourcesNASA NTRSDOE OSTI
DOE OSTI2015

Roles of individual radicals generated by a submerged dielectric barrier discharge plasma reactor during Escherichia coli O157:H7 inactivation

A submerged dielectric barrier discharge plasma reactor (underwater DBD) has been used on Escherichia coli O157:H7 (ATCC 35150). Plasma treatment was carried out using clean dry air gas to investigate the individual effects of the radicals produced by underwater DBD on an E. coli O157:H7 suspension (8.0 log CFU/ml). E. coli O157:H7 was reduced by 6.0 log CFU/ml for 2 min of underwater DBD plasma treatment. Optical Emission Spectra (OES) shows that OH and NO (α, β) radicals, generated by underwater DBD along with ozone gas. E. coli O157:H7 were reduced by 2.3 log CFU/ml for 10 min of underwater DBD plasma treatment with the terephthalic acid (TA) OH radical scavenger solution, which is significantly lower (3.7 log CFU/ml) than the result obtained without using the OH radical scavenger. A maximum of 1.5 ppm of ozone gas was produced during the discharge of underwater DBD, and the obtained reduction difference in E.coli O157:H7 in presence and in absence of ozone gas was 1.68 log CFU/ml. The remainder of the 0.62 log CFU/ml reduction might be due to the effect of the NO (α, β) radicals or due to the combined effect of all the radicals produced by underwater DBD. A small amount of hydrogen peroxide was also generated but does not play any role in E. coli O157:H7 inactivation.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2013

CATALOG AND STATISTICAL STUDY OF X-RAY SELECTED BL LACERTAE OBJECTS

This paper presents a catalog of 312 X-ray selected BL Lacerate objects (XBLs), optically identified through the end of 2011. It contains the names from different surveys, equatorial coordinates, redshifts, multifrequency flux values, and luminosities for each source. In addition, the different characteristics of XBLs are statistically investigated (redshift, radio/optical/X-ray luminosities, central black hole (BH) mass, synchrotron peak frequency, broadband spectral indices, optical flux variability). Their values are collected through an extensive bibliographic and database search or calculated by us. The redshifts range from 0.031 to 0.702 with a maximum of the distribution at z = 0.223. The 1.4 GHz luminosities of XBLs log {nu}L{sub {nu}} {approx} 39-42 erg s{sup -1} while optical V and X-ray 0.1-2.4 keV bands show log {nu}L{sub {nu}} {approx} 43-46 erg s{sup -1}. The XBL hosts are elliptical galaxies with effective radii r{sub eff} = 3.2625.40 kpc and ellipticities, in = 0.040.52. Their R-band absolute magnitudes M{sub R} range from -21.11 mag to -24.86 mag with a mean value of -22.83 mag. The V - R indices of the hosts span from 0.61 to 1.52 and reveal a fourth-degree polynomial relationship with z that enables us to evaluate the redshifts of five sources whose V - R indices were determined from the observations but whose irredshifts values are either not found or not confirmed. The XBL nuclei show a wider range of 7.31 mag for M{sub R} with the highest luminosity corresponding to M{sub R} = -27.24 mag. The masses of central BHs are found in the interval log M{sub BH} = 7.39-9.30 solar masses (with distribution maximum at log M{sub BH}/M{sub Sun} = 8.30). The synchrotron peak frequencies are spread over the range log {nu}{sub peak} = 14.56-19.18 Hz with a peak of the distribution at log {nu}{sub peak} = 16.60 Hz. The broadband radio-to-optical ({alpha}{sub ro}), optical-to-X-ray ({alpha}{sub ox}), and radio-to-X-ray ({alpha}{sub rx}) spectral indices are distributed in the intervals (0.17,0.59), (0.56,1.48), and (0.41,0.75), respectively. In the optical energy range, the overall flux variability increases, on average, towards shorter wavelengths: ({Delta}m) = 1.22, 1.50, and 1.82 mag through the R, V, B bands of Johnson-Cousins system, respectively. XBLs seem be optically less variable at the intranight timescales compared to the radio-selected BL Lacs (RBLs).

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2013

WILSON-BAPPU EFFECT: EXTENDED TO SURFACE GRAVITY

In 1957, Wilson and Bappu found a tight correlation between the stellar absolute visual magnitude (M{sub V} ) and the width of the Ca II K emission line for late-type stars. Here, we revisit the Wilson-Bappu relationship (WBR) to claim that the WBR can be an excellent indicator of stellar surface gravity of late-type stars as well as a distance indicator. We have measured the width (W) of the Ca II K emission line in high-resolution spectra of 125 late-type stars obtained with the Bohyunsan Optical Echelle Spectrograph and adopted from the Ultraviolet and Visual Echelle Spectrograph archive. Based on our measurement of the emission line width (W), we have obtained a WBR of M{sub V} = 33.76 - 18.08 log W. In order to extend the WBR to being a surface gravity indicator, stellar atmospheric parameters such as effective temperature (T{sub eff}), surface gravity (log g), metallicity ([Fe/H]), and micro-turbulence ({xi}{sub tur}) have been derived from self-consistent detailed analysis using the Kurucz stellar atmospheric model and the abundance analysis code, MOOG. Using these stellar parameters and log W, we found that log g = -5.85 log W+9.97 log T{sub eff} - 23.48 for late-type stars.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2011

AKARI OBSERVATIONS OF BROWN DWARFS. II. CO{sub 2} AS PROBE OF CARBON AND OXYGEN ABUNDANCES IN BROWN DWARFS

Recent observations with the infrared astronomical satellite AKARI have shown that the CO{sub 2} bands at 4.2 {mu}m in three brown dwarfs are much stronger than expected from the unified cloudy model (UCM) based on recent solar C and O abundances. This result has been a puzzle, but we now find it is simply due to the effect of C and O abundances. We show that these strong CO{sub 2} bands can be explained with the UCMs based on the classical C and O abundances (log A{sub C} and log A{sub O}), which are about 0.2 dex larger compared to the recent values. Since three other brown dwarfs could be interpreted fairly well with the recent solar C and O abundances, we require at least two model sequences based on the different chemical compositions to interpret all the AKARI spectra. The reason is that the CO{sub 2} band is especially sensitive to C and O abundances, since the CO{sub 2} abundance depends approximately on A{sub C} A{sup 2}{sub O}-the cube of C and O abundances. For this reason, even low-resolution spectra of very cool dwarfs, especially of CO{sub 2}, cannot be understood unless a model with proper abundances is applied. For the same reason, CO{sub 2} is an excellent indicator of C and O abundances, and we can now estimate C and O abundances of brown dwarfs as follows: Three of the six brown dwarfs observed with AKARI should have high C and O abundances similar to the classical solar values (e.g., log A{sub C} = 8.60 and log A{sub O} = 8.92), but the other three may have low C and O abundances similar to the recent solar values (e.g., log A{sub C} = 8.39 and log A{sub O} = 8.69). This result implies that three of the six brown dwarfs are highly metal-rich relative to the Sun, if the recent solar C and O abundances are correct.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2011

TWO TYPES OF EXTREME-ULTRAVIOLET BRIGHTENINGS IN AR 10926 OBSERVED BY HINODE/EIS

We have investigated seven extreme-ultraviolet (EUV) brightenings in the active region AR 10926 on 2006 December 2 observed by the EUV Imaging Spectrometer on board the Hinode spacecraft. We have determined their Doppler velocities and non-thermal velocities from 15 EUV spectral lines (log T = 4.7 - 6.4) by fitting each line profile to a Gaussian function. The Doppler velocity maps for different temperatures are presented to show the height dependence of the Doppler shifts. It is found that the active region brightenings show two distinct Doppler shift patterns. The type 1 brightening shows a systematic increase of Doppler velocity from -68 km s{sup -1} (strong blueshift) at log T = 4.7 to -2 km s{sup -1} (weak blueshift) at log T = 6.4, while the type 2 brightenings have Doppler velocities in the range from -20 km s{sup -1} to 20 km s{sup -1}. The type 1 brightening point is considered to sit in an upward reconnection outflow whose speed decreases with height. In both types of brightenings, the non-thermal velocity is found to be significantly enhanced at log T = 5.8 compared to the background region. We have also determined electron densities from line ratios and derived temperatures from emission measure loci using the CHIANTI atomic database. The electron densities of all brightenings are comparable to typical values in active regions (log N{sub e} = 9.9-10.4). The emission measure loci plots indicate that these brightenings should be multi-thermal whereas the background is isothermal. The differential emission measure as a function of temperature shows multiple peaks in the EUV brightening regions, while it has only a single peak (log T = 6.0) in the background region. Using Michelson Doppler Imager magnetograms, we have found that the type 1 brightening is associated with a canceling magnetic feature with a flux canceling rate of 2.4 x 10{sup 18} Mx hr{sup -1}. We also found the canceling magnetic feature and chromospheric brightenings in the type 1 brightening from the Hinode SOT and Transition Region and Coronal Explorer data. This observation corroborates our argument that brightening is caused by magnetic reconnection in a low atmosphere.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2012

THE 60 MONTH ALL-SKY BURST ALERT TELESCOPE SURVEY OF ACTIVE GALACTIC NUCLEUS AND THE ANISOTROPY OF NEARBY AGNs

Surveys above 10 keV represent one of the best resources to provide an unbiased census of the population of active galactic nuclei (AGNs). We present the results of 60 months of observation of the hard X-ray sky with Swift/Burst Alert Telescope (BAT). In this time frame, BAT-detected (in the 15-55 keV band) 720 sources in an all-sky survey of which 428 are associated with AGNs, most of which are nearby. Our sample has negligible incompleteness and statistics a factor of {approx}2 larger over similarly complete sets of AGNs. Our sample contains (at least) 15 bona fide Compton-thick AGNs and 3 likely candidates. Compton-thick AGNs represent {approx}5% of AGN samples detected above 15 keV. We use the BAT data set to refine the determination of the log N-log S of AGNs which is extremely important, now that NuSTAR prepares for launch, toward assessing the AGN contribution to the cosmic X-ray background. We show that the log N-log S of AGNs selected above 10 keV is now established to {approx}10% precision. We derive the luminosity function of Compton-thick AGNs and measure a space density of 7.9{sup +4.1}{sub -2.9} Multiplication-Sign 10{sup -5} Mpc{sup -3} for objects with a de-absorbed luminosity larger than 2 Multiplication-Sign 10{sup 42} erg s{sup -1}. As the BAT AGNs are all mostly local, they allow us to investigate the spatial distribution of AGNs in the nearby universe regardless of absorption. We find concentrations of AGNs that coincide spatially with the largest congregations of matter in the local ({<=}85 Mpc) universe. There is some evidence that the fraction of Seyfert 2 objects is larger than average in the direction of these dense regions.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2012

THE PROPERTIES OF TWO LOW-REDSHIFT O VI ABSORBERS AND THEIR ASSOCIATED GALAXIES TOWARD 3C 263 ,

Ultraviolet observations of the QSO 3C 263 (z{sub em} = 0.652) with Cosmic Origins Spectrograph and FUSE reveal O VI absorption systems at z = 0.06342 and 0.14072. WIYN multi-object spectrograph observations provide information about the galaxies associated with the absorbers. The multi-phase system at z = 0.06342 traces cool photoionized gas and warm collisionally ionized gas associated with an L {approx} 0.31 L* compact spiral emission line galaxy with an impact parameter of 63 kpc. The cool photoionized gas in the absorber is well modeled, with log U {approx} -2.6, log N(H) {approx} 17.8, log n(H) {approx} -3.3 and [Si/H] = -0.14 {+-} 0.23. The collisionally ionized gas containing C IV and O VI probably arises in cooling shock-heated transition temperature gas with log T {approx} 5.5. The absorber is likely tracing circumgalactic gas enriched by gas ejected from the spiral emission line galaxy. The simple system at z = 0.14072 only contains O VI and broad and narrow H I. The O VI with b = 33.4 {+-} 11.9 km s{sup -1} is likely associated with the broad H I {lambda}1215 absorption, with b = 86.7 {+-} 15.4 km s{sup -1}. The difference in Doppler parameters implies the detection of a very large column of warm gas with log T = 5.61(+0.16, -0.25), log N(H) = 19.54(+0.26, -0.44), and [O/H] = -1.48 (+0.46, -0.26). This absorber is possibly associated with a 1.6 L* absorption line galaxy with an impact parameter of 617 kpc, although an origin in warm filament gas or in the halo of a fainter galaxy is more likely.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2013

REGULATION OF BLACK HOLE WINDS AND JETS ACROSS THE MASS SCALE

We present a study of the mechanical power generated by both winds and jets across the black hole mass scale. We begin with the study of ionized X-ray winds and present a uniform analysis using Chandra grating spectra. The high-quality grating spectra facilitate the characterization of the outflow velocity, ionization, and column density of the absorbing gas. We find that the kinetic power of the winds, derived from these observed quantities, scales with increasing bolometric luminosity as log (L {sub wind,42}/C{sub v} ) = (1.58 {+-} 0.07)log (L {sub Bol,42}) - (3.19 {+-} 0.19). This suggests that supermassive black holes may be more efficient than stellar-mass black holes in launching winds, per unit filling factor, C{sub v} . If the black hole binary (BHB) and active galactic nucleus (AGN) samples are fit individually, the slopes flatten to {alpha}{sup BHB} = 0.91 {+-} 0.31 and {alpha}{sup AGN} = 0.63 {+-} 0.30 (formally consistent within errors). The broad fit and individual fits both characterize the data fairly well, and the possibility of common slopes may point to common driving mechanisms across the mass scale. For comparison, we examine jet production, estimating jet power based on the energy required to inflate local bubbles. The jet relation is log (L {sub Jet,42}) = (1.18 {+-} 0.24)log (L {sub Bondi,42}) - (0.96 {+-} 0.43). The energetics of the bubble associated with Cygnus X-1 are particularly difficult to determine, and the bubble could be a background supernova remnant. If we exclude Cygnus X-1 from our fits, then the jets follow a relation consistent with the winds, but with a higher intercept, log (L {sub Jet,42}) = (1.34 {+-} 0.50)log (L {sub Bondi,42}) - (0.80 {+-} 0.82). The formal consistency in the wind and jet scaling relations, when assuming that L {sub Bol} and L {sub Bondi} are both proxies for mass accretion rate, suggests that a common launching mechanism may drive both flows; magnetic processes, such as magnetohydrodynamics and magnetocentrifugal forces, are viable possibilities. We also examine winds that are moving at especially high velocities, v > 0.01c. These ultra-fast outflows tend to resemble the jets more than the winds in terms of outflow power, indicating that we may be observing a regime in which winds become jets. A transition at approximately L {sub Bol} Almost-Equal-To 10{sup -2} L {sub Edd} is apparent when outflow power is plotted versus Eddington fraction. At low Eddington fractions, the jet power is dominant, and at high Eddington fractions, the wind power is dominant. This study allows for the total power from black hole accretion, both mechanical and radiative, to be characterized in a simple manner and suggests possible connections between winds and jets. X-ray wind data and jet cavity data will enable stronger tests.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

SDSS J074511.56+194926.5: Discovery of a metal-rich and tidally distorted extremely low mass white dwarf

We present the discovery of an unusual, tidally distorted extremely low mass white dwarf (WD) with nearly solar metallicity. Radial velocity measurements confirm that this is a compact binary with an orbital period of 2.6975 hr and a velocity semi-amplitude of K = 108.7 km s{sup –1}. Analysis of the hydrogen Balmer lines yields an effective temperature of T {sub eff} = 8380 K and a surface gravity of log g = 6.21 that in turn indicate a mass of M = 0.16 M {sub ☉} and a cooling age of 4.2 Gyr. In addition, a detailed analysis of the observed metal lines yields abundances of log (Mg/H) = –3.90, log (Ca/H) = –5.80, log (Ti/H) = –6.10, log (Cr/H) = –5.60, and log (Fe/H) = –4.50, similar to the sun. We see no evidence of a debris disk from which these metals would be accreted, though the possibility cannot entirely be ruled out. Other potential mechanisms to explain the presence of heavy elements are discussed. Finally, we expect this system to ultimately undergo unstable mass transfer and merge to form a ∼0.3-0.6 M {sub ☉} WD in a few Gyr.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2014

SDSS/SEGUE spectral feature analysis for stellar atmospheric parameter estimation

Large-scale and deep sky survey missions are rapidly collecting a large amount of stellar spectra, which necessitate the estimation of atmospheric parameters directly from spectra and make it feasible to statistically investigate latent principles in a large data set. We present a technique for estimating parameters T{sub eff}, log g, and [Fe/H] from stellar spectra. With this technique, we first extract features from stellar spectra using the LASSO algorithm; then, the parameters are estimated from the extracted features using the support vector regression. On a subsample of 20,000 stellar spectra from the Sloan Digital Sky Survey (SDSS) with reference parameters provided by the SDSS/SEGUE Spectroscopic Parameter Pipeline, estimation consistency are 0.007458 dex for log T{sub eff} (101.609921 K for T{sub eff}), 0.189557 dex for log g, and 0.182060 for [Fe/H], where the consistency is evaluated by mean absolute error. Prominent characteristics of the proposed scheme are sparseness, locality, and physical interpretability. In this work, each spectrum consists of 3821 fluxes, and 10, 19, and 14 typical wavelength positions are detected, respectively, for estimating T{sub eff}, log g, and [Fe/H]. It is shown that the positions are related to typical lines of stellar spectra. This characteristic is important in investigating physical indications from analysis results. Then, stellar spectra can be described by the individual fluxes on the detected positions (PD) or local integration of fluxes near them (LI). The aforementioned consistency is the result based on features described by LI. If features are described by PD, consistency is 0.009092 dex for log T{sub eff} (124.545075 K for T{sub eff}), 0.198928 dex for log g, and 0.206814 dex for [Fe/H].

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2015

A DEFINITION FOR GIANT PLANETS BASED ON THE MASS–DENSITY RELATIONSHIP

We present the mass–density relationship (log M − log ρ) for objects with masses ranging from planets (M ≈ 0.01 M{sub Jup}) to stars (M > 0.08 M{sub ⊙}). This relationship shows three distinct regions separated by a change in slope in the log M − log ρ plane. In particular, objects with masses in the range 0.3 M{sub Jup}–60 M{sub Jup} follow a tight linear relationship with no distinguishing feature to separate the low-mass end (giant planets) from the high-mass end (brown dwarfs). We propose a new definition of giant planets simply based on changes in the slope of the log M versus log ρ relationship. By this criterion, objects with masses less than ≈0.3 M{sub Jup} are low-mass planets, either icy or rocky. Giant planets cover the mass range 0.3 M{sub Jup}–60 M{sub Jup}. Analogous to the stellar main sequence, objects on the upper end of the giant planet sequence (brown dwarfs) can simply be referred to as “high-mass giant planets,” while planets with masses near that of Jupiter can be called “low-mass giant planets.”.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Sparse Cholesky factorization for solving nonlinear PDEs via Gaussian processes

In recent years, there has been widespread adoption of machine learning-based approaches to automate the solving of partial differential equations (PDEs). Among these approaches, Gaussian processes (GPs) and kernel methods have garnered considerable interest due to their flexibility, robust theoretical guarantees, and close ties to traditional methods. They can transform the solving of general nonlinear PDEs into solving quadratic optimization problems with nonlinear, PDE-induced constraints. However, the complexity bottleneck lies in computing with dense kernel matrices obtained from pointwise evaluations of the covariance kernel, and its partial derivatives, a result of the PDE constraint and for which fast algorithms are scarce. The primary goal of this paper is to provide a near-linear complexity algorithm for working with such kernel matrices. We present a sparse Cholesky factorization algorithm for these matrices based on the near-sparsity of the Cholesky factor under a novel ordering of pointwise and derivative measurements. The near-sparsity is rigorously justified by directly connecting the factor to GP regression and exponential decay of basis functions in numerical homogenization. We then employ the Vecchia approximation of GPs, which is optimal in the Kullback-Leibler divergence, to compute the approximate factor. This enables us to compute ϵ-approximate inverse Cholesky factors of the kernel matrices with complexity O(N log d (N/ϵ)) in space and O(N log 2d (N/ϵ)) in time. We integrate sparse Cholesky factorizations into optimization algorithms to obtain fast solvers of the nonlinear PDE. We numerically illustrate our algorithm’s near-linear space/time complexity for a broad class of nonlinear PDEs such as the nonlinear elliptic, Burgers, and Monge-Ampère equations. In summary, we provide a fast, scalable, and accurate method for solving general PDEs with GPs and kernel methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

SHαDE: survey description and mass–kinematics scaling relations for dwarf galaxies

ABSTRACT The Study of H α from Dwarf Emissions (SHαDE) is a high spectral resolution (R = 13 500) H α integral field survey of 69 dwarf galaxies with stellar masses 106 < M⋆ < 109 M⊙. The survey used FLAMES on the ESO Very Large Telescope. SHαDE is designed to study the kinematics and stellar populations of dwarf galaxies using consistent methods applied to massive galaxies and at matching level of detail, connecting these mass ranges in an unbiased way. In this paper, we set out the science goals of SHαDE, describe the sample properties, outline the data reduction, and analysis processes. We investigate the log M⋆–log S0.5 mass–kinematics scaling relation, which has previously shown potential for combining galaxies of all morphologies in a single scaling relation. We extend the scaling relation from massive galaxies to dwarf galaxies, demonstrating this relation is linear down to a stellar mass of M⋆ ∼ 108.6 M⊙. Below this limit, the kinematics of galaxies inside one effective radius appears to be dominated by the internal velocity dispersion limit of the H α-emitting gas, giving a bend in the log M⋆–log S0.5 relation. Replacing stellar mass with total baryonic mass using gas mass estimate reduces the severity but does not remove the linearity limit of the scaling relation. An extrapolation to estimate the galaxies’ dark matter halo masses, yields a log Mh–log S0.5 scaling relation that is free of any bend, has reduced curvature over the whole mass range, and brings galaxies of all masses and morphologies on to the virial relation.

Barat, Dilyar↗
DOE OSTI2023

Abundances of CNO elements in z ∼ 0.3–0.4 Lyman continuum leaking galaxies

We present observations with the Space Telescope Imaging Spectrograph onboard the Hubble Space Telescope (HST) of 11 Lyman continuum (LyC) leaking galaxies at redshifts, z , in the range 0.29–0.43, with oxygen abundances 12+log(O/H) = 7.64–8.16, stellar masses M ⋆ ∼ 10 7.8 –10 9.8 M ⊙ , and O 32 = [O III ] λ5007/[O II ] λ3727 of ∼ 5–20 aiming to detect the C III ] λ1908 emission line. We combine these observations with the optical Sloan Digital Sky Survey (SDSS) spectra for the determination of carbon, nitrogen, and oxygen abundances. Our sample was supplemented by 31 galaxies from the literature, for which carbon, nitrogen, and oxygen abundances can be derived from the HST and SDSS spectra. These additional galaxies, however, do not have LyC observations. We find that log(C/O) for the entire sample at 12+log(O/H) < 8.1 does not depend on metallicity, with a small dispersion of ∼0.13 dex around the average value of ∼−0.75 dex. On the other hand, the log(N/O) in galaxies at z > 0.1, including LyC leakers, is systematically higher compared to the rest of the sample with lower metallicity. We find that log(C/O) slightly decreases with increasing M ⋆ from ∼ −0.65 at M ⋆ = 10 6 M ⊙ to ∼ −0.80 at M ⋆ = 10 9 –10 10 M ⊙ , whereas log(N/O) is considerably enhanced at M ⋆ > 10 8 M ⊙ . The origin of these trends remains basically unknown. A possible solution would be to assume that the upper mass limit of the stellar initial mass function in more massive galaxies is higher. This would result in a higher production of oxygen and a larger fraction of massive stars with stellar wind polluting the interstellar medium with nitrogen.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2023

A comparison of the baryonic Tully–Fisher relation in MaNGA and IllustrisTNG

We compare an observed baryonic Tully–Fisher relation (BTFR) from the Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) and HI-MaNGA surveys to a simulated BTFR from the cosmological magnetohydrodynamical simulation, IllustrisTNG. To do so, we calibrate the BTFR of the local universe using 377 galaxies from the MaNGA and HI-MaNGA surveys, and perform mock 21 cm observations of matching galaxies from IllustrisTNG. The mock observations are used to ensure that the comparison with the observed galaxies is fair since it has identical measurement algorithms, observational limitations, biases, and uncertainties. For comparison, we also calculate the BTFR for the simulation without mock observations and demonstrate how mock observations are necessary to fairly and consistently compare between observational and theoretical data. We report a MaNGA BTFR of log$_{10} (M_{ \rm Bary}/M_\odot)= (2.97 \pm 0.18)$ log$_{10} V_{ \rm Rot} + (4.04 \pm 0.41)\, \log _{10}{M_{\odot }}$ and an IllustrisTNG BTFR of log$_{10} (M_{ \rm Bary}/M_\odot) = (2.94 \pm 0.23$) log$_{10} V_{ \rm Rot} + (4.15 \pm 0.44)\, \log _{10}{M_{\odot }}$. Thus, MaNGA and IllustrisTNG produce BTFRs that agree within uncertainties, demonstrating that IllustrisTNG has created a galaxy population that obeys the observed relationship between mass and rotation velocity in the observed universe.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2021

Sentiment Analysis based Error Detection for Large-Scale Systems

Today's large-scale systems such as High Performance Computing (HPC) Systems are designed/utilized towards exascale computing, inevitably decreasing its reliability due to the increasing design complexity. HPC systems conduct extensive logging of their execution behaviour. In this paper, we leverage the inherent meaning behind the log messages and propose a novel sentiment analysis-based approach for the error detection in large-scale systems, by automatically mining the sentiments in the log messages. Our contributions are four-fold. (1) We develop a machine learning (ML) based approach to automatically build a sentiment lexicon, based on the system log message templates. (2) Using the sentiment lexicon, we develop an algorithm to detect system errors. (3) We develop an algorithm to identify the nodes and components with erroneous behaviors, based on sentiment polarity scores. (4) We evaluate our solution vs. other state-of-the-art machine/deep learning algorithms based on three representative supercomputers' system logs. Experiments show that our error detection algorithm can identify error messages with an average MCC score and f -score of 91% and 96% respectively, while state of the art ML/deep learning model (LSTM) obtains only 67% and 84%. To the best of our knowledge, this is the first work leveraging the sentiments embedded in log entries of large-scale systems for system health analysis.

error detection↗
DOE OSTI2018

Spatio-Temporal Analysis of HPC I/O and Connection Data

The HPC system consists of a set of layers of software and hardware for I/O and networking. System logs are helpful resources to understand what is going on in the system. A challenge is that it is non-trivial to analyze the logs maintained in various levels of the stack. Independent analysis might lead to an incomplete conclusion due to the limited coverage of each log. This work takes a comprehensive approach to analysis that incorporates the logs in the multiple layers and components, in order to facilitate the detection of anomalous activities. This research aims to identify and predict potential performance bottlenecks in the HPC system, by capturing the temporal variation patterns from heterogeneous, high-dimensional, and non-linear log data. In this paper, we share our preliminary efforts for spatio-temporal analysis of HPC I/O and connection data, with our initial observations from the analysis of one-week HPC log data sets collected from one of NERSC systems.

Choi, Jinhwan↗
DOE OSTI2016

SU-G-JeP1-08: Dual Modality Verification for Respiratory Gating Using New Real- Time Tumor Tracking Radiotherapy System

Purpose: The respirato ry gating system combined the TrueBeam and a new real-time tumor-tracking radiotherapy system (RTRT) was installed. The RTRT system consists of two x-ray tubes and color image intensifiers. Using fluoroscopic images, the fiducial marker which was implanted near the tumor was tracked and was used as the internal surrogate for respiratory gating. The purposes of this study was to develop the verification technique of the respiratory gating with the new RTRT using cine electronic portal image device images (EPIDs) of TrueBeam and log files of the RTRT. Methods: A patient who underwent respiratory gated SBRT of the lung using the RTRT were enrolled in this study. For a patient, the log files of three-dimensional coordinate of fiducial marker used as an internal surrogate were acquired using the RTRT. Simultaneously, the cine EPIDs were acquired during respiratory gated radiotherapy. The data acquisition was performed for one field at five sessions during the course of SBRT. The residual motion errors were calculated using the log files (E{sub log}). The fiducial marker used as an internal surrogate into the cine EPIDs was automatically extracted by in-house software based on the template-matching algorithm. The differences between the the marker positions of cine EPIDs and digitally reconstructed radiograph were calculated (E{sub EPID}). Results: Marker detection on EPID using in-house software was influenced by low image contrast. For one field during the course of SBRT, the respiratory gating using the RTRT showed the mean ± S.D. of 95{sup th} percentile E{sub EPID} were 1.3 ± 0.3 mm,1.1 ± 0.5 mm,and those of E{sub log} were 1.5 ± 0.2 mm, 1.1 ± 0.2 mm in LR and SI directions, respectively. Conclusion: We have developed the verification method of respiratory gating combined TrueBeam and new real-time tumor-tracking radiotherapy system using EPIDs and log files.

60 APPLIED LIFE SCIENCES↗