Engineering Papers⌕ Search

Engineering topics

Jones, Mark

Publications and source records attributed to Jones, Mark.

At least 19 records

Decode the Workload: Training Deep Learning Models for Efficient Compute Cluster Representation

Monitoring the status of a high throughput computing cluster running computationally intensive production jobs is a crucial yet challenging system administration task due to the complexity of such systems. To this end, we train autoencoders using the Linux kernel CPU metrics of the cluster. Additionally, we explore assisting these models with graph neural networks to share information across threads within a compute node. The models are compared in terms of their ability to: 1) Produce a compressed latent representation that captures the salient features of the input, 2) Detect anomalous activity, and 3) Make distinction between different kinds of jobs run at Jefferson Lab. The goal is to have a robust encoder whose compressed embeddings are used for several downstream tasks. We extend this study further by deploying these models in a human-in-the-loop production-based setting for the anomaly detection task and discuss the associated implementation aspects such as continual learning and the criterion to generate alarms. This study represents a first step in the endeavor towards building self-supervised large-scale foundation models for computing centers.

Mohammed, Ahmed↗

Detection and characterization of detached tidal dwarf galaxies

Tidal interactions between galaxies often give rise to tidal tails, which can harbor concentrations of stars and interstellar gas resembling dwarf galaxies. Some of these tidal dwarf galaxies (TDGs) have the potential to detach from their parent galaxies and become independent entities, but their long-term survival is uncertain. In this study, we conducted a search for detached TDGs associated with a sample of 39 interacting galaxy pairs in the local Universe using infrared, ultraviolet, and optical images. We employed IR colors and UV/optical/IR spectral energy distributions to identify potential interlopers, such as foreground stars or background quasars. Through spectroscopic observations using the Boller and Chivens spectrograph at San Pedro Mártir Observatory, we confirmed that six candidate TDGs are at the same redshift as their putative parent galaxy pairs. We identified and measured emission lines in the optical spectra and calculated nebular oxygen abundances, which range from log(O/H) = 8.10 ± 0.01 to 8.51 ± 0.02. We have serendipitously discovered an additional detached TDG candidate in Arp72 using available spectra from SDSS. Utilizing the photometric data and the CIGALE code for stellar population and dust emission fitting, we derived the stellar masses, stellar population ages, and stellar metallicities for these detached TDGs. Compared to standard mass-metallicity relations for dwarf galaxies, five of the seven candidates have higher than expected metallicities, confirming their tidal origins. One of the seven candidates remains unclear due to large uncertainties in metallicity, and another has stellar and nebular metallicities compatible with those of a preexisting dwarf galaxy. The latter object is relatively compact in the optical relative to its stellar mass, in contrast to the other candidate TDGs, which have large diameters for their stellar masses compared to most dwarf galaxies. The derived stellar population ages range from 100 Myr to 900 Myr, while the inferred stellar masses are between 2 × 10 6 M ⊙ and 8 × 10 7 M ⊙ . Four of the six TDGs are associated with the gas-rich M51-like pair Arp 72, one TDG is associated with a second M51-like pair Arp 86, and another is associated with Arp 65, an approximately equal mass pair. In spite of the relatively low stellar masses of these TDGs, they have survived for at least 100–900 Myrs, suggesting that they are stable and in dynamical equilibrium. We conclude that encounters with a relatively low-mass companion (1/10th–1/4th of the mass of the primary) can also produce long-lasting TDGs.

Astronomy & Astrophysics↗

Investigating Anomalies in Compute Clusters: An Unsupervised Learning Approach

As compute clusters continue to grow in scale and complexity, the frequency of detected anomalies in their operation significantly increases. Timely detection of anomalous events is vital to maintain system efficiency and availability. This study presents an attentionbased graph neural network (GNN) for detecting anomalies in clusters at the compute node level and for providing detailed root cause analysis. We show the effectiveness of attention-based GNNs to accurately detect and localize anomalies on real-world datasets.

Lu, Yiyang↗

The Suisun Summit 2022: Emerging Themes and Open Questions in qEEG and Neurotherapies

The Suisun Summit 2022 was a gathering of 70 clinicians, educators, and researchers in quantitative electroencephalography and neurofeedback. During this 5-day event, several themes emerged in talks or discussion groups: EEG/qEEG Reading Skills; Medication Effects and Pharmaco-EEG; Technological Advancements; Emerging Concerns; and Growing Community Prestige through Research. Participants were asked to summarize what they believed to be the most important messages from the event to share with colleagues who were not in attendance, resulting in this review. A unifying concept for all the themes was a desire for higher quality, standardized EEG/qEEG education that provides depth as well as breadth. Models of clinical care that encourage open communication with prescribers and functional medicine specialists were strongly emphasized. Abstracts from all presentations are attached in Addendum B.

60 APPLIED LIFE SCIENCES↗

Multi-mode analysis of surface losses in a superconducting microwave resonator in high magnetic fields

This paper reports on a surface impedance measurement of a bulk metal niobium–titanium superconducting radio frequency (SRF) cavity in a magnetic field (up to 10 T). Here, a novel method is employed to decompose the surface resistance contributions of the cylindrical cavity end caps and walls using measurements from multiple TM cavity modes. The results confirm that quality factor degradation of a NbTi SRF cavity in a high magnetic field is primarily from surfaces perpendicular to the field (the cavity end caps), while parallel surface resistances (the walls) remain relatively constant. This result is encouraging for applications needing high Q cavities in strong magnetic fields, such as the Axion Dark Matter eXperiment because it opens the possibility of hybrid SRF cavity construction to replace conventional copper cavities.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Trigeminal Neuralgia Treated With Stereotactic Radiosurgery: The Effect of Dose Escalation on Pain Control and Treatment Outcomes

Purpose: To analyze the effect of dose escalation on treatment outcome in patients undergoing stereotactic radiosurgery (SRS) for trigeminal neuralgia (TN). Methods and Materials: A retrospective review was performed of 870 patients who underwent SRS for a diagnosis of TN from 2 institutions. Patients were typically treated using a single 4-mm isocenter placed at the trigeminal nerve dorsal root entry zone. Patients were divided into groups based on treatment doses: ≤82 Gy (352 patients), 83 to 86 Gy (85 patients), and ≥90 Gy (433 patients). Pain response was classified using a categorical scoring system, with fair or poor pain control representing treatment failure. Treatment-related facial numbness was classified using the Barrow Neurological Institute scale. Log-rank tests were performed to test differences in time to pain failure or development of facial numbness for patients treated with different doses. Results: Median age at first pain onset was 63 years, median age at time of SRS was 71 years, and median follow-up was 36.5 months from the time of SRS. A majority of patients (827, 95%) were clinically diagnosed with typical TN. The 4-year rate of excellent to good pain relief was 87% (95% confidence interval 84%-90%). The 4-year rate of pain response was 79%, 82%, and 92% in patients treated to ≤82 Gy, 83 to 86 Gy, and ≥90 Gy, respectively. Patients treated to doses ≤82 Gy had an increased risk of pain failure after SRS, compared with patients treated to ≥90 Gy (hazard ratio 2.0, P=.0007). Rates of treatment-related facial numbness were similar among patients treated to doses ≥83 Gy. Nine patients (1%) were diagnosed with anesthesia dolorosa. Conclusions: Dose escalation for TN to doses >82 Gy is associated with an improvement in response to treatment and duration of pain relief. Patients treated at these doses, however, should be counseled about the increased risk of treatment-related facial numbness.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Resolution Performance of HERA-B Lead-Glass Calorimeters

In preparation of upcoming 12GeV experiments at Thomas Jefferson National Accelerator Facility it is necessary to upgrade existing systems or install new detectors. As part of this effort, an array of lead-glass sampling calorimeters is need for use in the GEP-5 experiment. A sampling calorimeter can be used to determine the energy and spatial position of a high energy particle that enters it while simultaneously stopping the particle. To determine the appropriate construction to meet the needs of upcoming experiments, it was necessary to take an existing model and confirm its energy and position resolution. This model could then be confirmed as an option for the final construction, or used as a starting point to design a better detector. For our test we obtained ten lead-glass calorimeters used in HERA-B and tested them in End Station A at SLAC. I will report on our findings for the HERA-B lead-glass sampling calorimeters. I will cover the results of both the energy and position resolutions as well as the methods used to determine these quantities.

Losada, Anthony↗

PDRD (SR13046) Tritium Production Final Report

Utilizing the results of Texas A&M University (TAMU) senior design projects on tritium production in four different small modular reactors (SMR), the Savannah River National Laboratory’s (SRNL) developed an optimization model evaluating tritium production versus uranium utilization under a FY2013 plant directed research development (PDRD) project. The model is a tool that can evaluate varying scenarios and various reactor designs to maximize the production of tritium per unit of unobligated United States (US) origin uranium that is in limited supply.

07 ISOTOPE AND RADIATION SOURCES↗

Observation of the Λ ⁷He Hypernucleus by the (e, e'K⁺) Reaction

An experiment with a newly developed high-resolution kaon spectrometer and a scattered electron spectrometer with a novel configuration was performed in Hall C at Jefferson Lab. The ground state of a neutron-rich hypernucleus, He Λ ⁷, was observed for the first time with the (e, e'K⁺) reaction with an energy resolution of ~0.6 MeV. This resolution is the best reported to date for hypernuclear reaction spectroscopy. The He Λ ⁷ binding energy supplies the last missing information of the A=7, T=1 hypernuclear isotriplet, providing a new input for the charge symmetry breaking effect of the ΛN potential.

Nakamura, S. N.↗

Measurement of the Spin Asymmetries of the Proton

The Spin Asymmetries of the Nucleon Experiment (SANE) measured the parallel, $A_\parallel$, and near-perpendicular, $A_{80}$, double spin asymmetries of the proton in inclusive electron scattering. Longitudinally polarized electrons were scattered from an ammonia target which provided polarized protons via dynamic nuclear polarization. The scattered electrons were detected using the Big Electron Telescope Array, BETA, which was centered at 40$^\circ$ and covered a large solid angle. BETA consisted of a scintillator hodoscope, gas Cherenkov, lucite hodoscope and a large array of lead glass detectors. The data was taken at the Thomas Jefferson National Accelerator Facility's Hall C at beam energies of $4.7$ and $5.9~GeV$ and covered $2.5~GeV^2 < Q^2 < 6.5~GeV^2$. From the $A_\parallel$ and $A_{80}$ measurements, the proton's spin asymmetries, $A_1$ and $A_2$, and the spin structure functions of the proton, $g_1$ and $g_2$, can be extracted. The measurements are in a range of Bjorken $x$, $0.3 < x < 0.8$, where extraction of the twist three matrix element $d_2^p $ (an integral of $g_1$ and $g_2$ weighted by $x^2$) is most sensitive.

Jones, Mark↗

Probing the High Momentum Component of the Deuteron at High Q{sup 2}

The {sup 2}H(e,e{sup '}p)n cross section at a momentum transfer of 3.5 (GeV/c){sup 2} was measured over a kinematical range that made it possible to study this reaction for a set of fixed missing momenta as a function of the neutron recoil angle {theta}{sub nq} and to extract missing momentum distributions for fixed values of {theta}{sub nq} up to 0.55 GeV/c. In the region of 35 deg. {<=}{theta}{sub nq}{<=}45 deg. recent calculations, which predict that final-state interactions are small, agree reasonably well with the experimental data. Therefore, these experimental reduced cross sections provide direct access to the high momentum component of the deuteron momentum distribution in exclusive deuteron electrodisintegration.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Successful Implementation of Image-Guided Radiation Therapy Quality Assurance in the Trans Tasman Radiation Oncology Group 08.01 PROFIT Study

Purpose: The quality assurance (QA) of image-guided radiation therapy (IGRT) within clinical trials is in its infancy, but its importance will continue to grow as IGRT becomes the standard of care. The purpose of this study was to demonstrate the feasibility of IGRT QA as part of the credentialing process for a clinical trial. Methods and Materials: As part of the accreditation process for a randomized trial in prostate cancer hypofraction, IGRT benchmarking across multiple sites was incorporated. Each participating site underwent IGRT credentialing via a site visit. In all centers, intraprostatic fiducials were used. A real-time assessment of analysis of IGRT was performed using Varian's Offline Review image analysis package. Two-dimensional (2D) kV and MV electronic portal imaging prostate patient datasets were used, consisting of 39 treatment verification images for 2D/2D comparison with the digitally reconstructed radiograph derived from the planning scan. The influence of differing sites, image modality, and observer experience on IGRT was then assessed. Results: Statistical analysis of the mean mismatch errors showed that IGRT analysis was performed uniformly regardless of institution, therapist seniority, or imaging modality across the three orthogonal planes. Conclusions: The IGRT component of clinical trials that include sophisticated planning and treatment protocols must undergo stringent QA. The IGRT technique of intraprostatic fiducials has been shown in the context of this trial to be undertaken in a uniform manner across Australia. Extending this concept to many sites with different equipment and IGRT experience will require a robust remote credentialing process.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

Measurements of the Electric Form Factor of the Neutron up to Q2 = 3.4 GeV2 using the Reaction He-3(e,e&#x27;n)pp

The electric form factor of the neutron was determined from studies of the reaction \rea{} in quasi-elastic kinematics in Hall A at Jefferson Lab. Longitudinally polarized electrons were scattered off a polarized target in which the nuclear polarization was oriented perpendicular to the momentum transfer. The scattered electrons were detected in a magnetic spectrometer in coincidence with neutrons that were registered in a large-solid-angle detector. More than doubling the $Q^2$-range over which it is known, we find \GEn{}$ = 0.0225 \pm 0.0017 (stat) \pm 0.0024 (syst)$, $0.0200 \pm 0.0023 \pm 0.0018$, and $0.0142 \pm 0.0019 \pm 0.0013$ for $Q^2$ = 1.72, 2.48, and 3.41~\gevsq, respectively.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗