Search NASA⌕ Search

SEARCH · Search NASA

Results for “software performance”

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 1,063 records · Page 59

Safety assessment of historical masonry churches based on pre-assigned kinematic limit analysis, FE limit and pushover analyses

This study presents some results of a comprehensive numerical analysis on three masonry churches damaged by the recent Emilia-Romagna (Italy) seismic events occurred in May 2012. The numerical study comprises: (a) pushover analyses conducted with a commercial code, standard nonlinear material models and two different horizontal load distributions; (b) FE kinematic limit analyses performed using a non-commercial software based on a preliminary homogenization of the masonry materials and a subsequent limit analysis with triangular elements and interfaces; (c) kinematic limit analyses conducted in agreement with the Italian code and based on the a-priori assumption of preassigned failure mechanisms, where the masonry material is considered unable to withstand tensile stresses. All models are capable of giving information on the active failure mechanism and the base shear at failure, which, if properly made non-dimensional with the weight of the structure, gives also an indication of the horizontal peak ground acceleration causing the collapse of the church. The results obtained from all three models indicate that the collapse is usually due to the activation of partial mechanisms (apse, façade, lateral walls, etc.). Moreover the horizontal peak ground acceleration associated to the collapse is largely lower than that required in that seismic zone by the Italian code for ordinary buildings. These outcomes highlight that structural upgrading interventions would be extremely beneficial for the considerable reduction of the seismic vulnerability of such kind of historical structures.

42 ENGINEERING↗

Investigation of static and dynamic behavior of functionally graded piezoelectric actuated Poly-Si micro cantilever probe

In the present paper a novel functionally graded piezoelectric (FGP) actuated Poly-Si micro cantilever probe is proposed for atomic force microscope. The shear piezoelectric coefficient d{sub 15} has much higher value than coupling coefficients d{sub 31} and d{sub 33}, hence in the present work the micro cantilever beam actuated by d{sub 15} effect is utilized. The material properties are graded in the thickness direction of actuator by a simple power law. A three dimensional finite element analysis has been performed using COMSOL Multiphysics® (version 4.2) software. Tip deflection and free vibration analysis for the micro cantilever probe has been done. The results presented in the paper shall be useful in the design of micro cantilever probe and their subsequent utilization in atomic force microscopes.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Power accounting of plasma discharges in the linear device Proto-MPEX

Plasma material interaction (PMI) studies are crucial to the successful development of future fusion reactors. Prototype Material Plasma Exposure eXperiment (Proto-MPEX) is a prototype design for the MPEX, a steady-state linear device being developed to study PMI. The primary purpose of Proto-MPEX is developing the plasma heating source concepts for MPEX. A power accounting study of Proto-MPEX works to identify machine operating parameters that could improve its performance, thereby increasing its PMI research capabilities, potentially impacting the MPEX design concept. To build a comprehensive power balance, an analysis of the helicon region has been performed implementing a diagnostic suite and software modeling to identify mechanisms and locations of heat loss from the main plasma. Of the 106.3 kW of input power, up to 90.5% of the power has been accounted for in the helicon region. When the analysis was extended to encompass the device to its end plates, 49.2% of the input power was accounted for and verified diagnostically. Areas requiring further diagnostic analysis are identified. The required improvements will be implemented in future work. The data acquisition and analysis processes will be streamlined to form a working model for future power balance studies of Proto-MPEX.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Progress toward Accelogic compression in ROOT

For the last 7 years, Accelogic pioneered and perfected a radically new theory of numerical computing codenamed “Compressive Computing”, which has an extremely profound impact on real-world computer science [1]. At the core of this new theory is the discovery of one of its fundamental theorems which states that, under very general conditions, the vast majority (typically between 70% and 80%) of the bits used in modern large-scale numerical computations are absolutely irrelevant for the accuracy of the end result. This theory of Compressive Computing provides mechanisms able to identify (with high intelligence and surgical accuracy) the number of bits (i.e., the precision) that can be used to represent numbers without affecting the substance of the end results, as they are computed and vary in real time. The bottom-line outcome will be to provide state-of-the-art compression algorithms --and accompanying software libraries-- able to surpass the performance of the compression engines currently available in the ROOT [7] framework. The resulting technology has the capability to enable substantial economic and operational gains (including speedup) for High Energy and Nuclear Physics data storage/analysis. In our initial studies, a factor of nearly x4 (3.9) compression was achieved with RHIC/STAR data where ROOT compression managed only x1.4 [6].As a collaboration of experimental scientists, private industry, and the ROOT Team, our aim is to capitalize on the substantial success delivered by the initial effort and produce a robust technology properly packaged as an open-source tool that could be used by virtually every experiment around the world as means for improving data management and accessibility.In this contribution, we will present our efforts integrating our concepts of “functionally lossless compression” within the ROOT framework implementation, with the purpose of producing a basic solution readily integrated into HENP applications. We will also present our progress applying this compression through realistic examples of analysis from both the STAR and CMS experiments.

Canal, Ph.↗

Extrapolating W -associated jet-production ratios at the LHC

Electroweak vector-boson production, accompanied by multiple jets, is an important background to searches for physics beyond the standard model. A precise and quantitative understanding of this process is helpful in constraining deviations from known physics. We study four key ratios in W + n -jet production at the LHC. We compute the ratio of cross sections for W + n - to W + (n-1)-jet production as a function of the minimum jet transverse momentum. We also study the ratio differentially, as a function of the W-boson transverse momentum; as a function of the scalar sum of the jet transverse energy, H$jets\atop{T}$; and as a function of certain jet transverse momenta. We show how to use such ratios to extrapolate differential cross sections to W + 6 -jet production at next-to-leading order, and we cross-check the method against a direct calculation at leading order. We predict the differential distribution in H$jets\atop{T}$ for W + 6 jets at next-to-leading order using such an extrapolation. We use the BlackHat software library together with SHERPA to perform the computations.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Prometheus: Coherent Exploration of Hardware and Software Optimizations using Aspen

With the dramatic increase in scale expected for Exascale computing, there is a dire need for tuning of hardware configurations and software optimizations such that they are in unison. However, the expected increase in tunable hardware parameters makes searching through the design space for optimal hardware-and-software configurations much more challenging. Towards this end, we propose a composable hardware-software optimization framework called Prometheus. Prometheus uses a combination of analytical and machine-learning techniques to capture application characteristics and subsequently determine the hardware-software configuration for near-optimal performance. We evaluate Prometheus for its efficacy using four widely used proxy applications: LULESH, CoMD, CG and CoEVP. We demonstrate that Prometheus identifies near-optimal hardwaresoftware configurations and verify the results via brute-force search of the design space.

Umar, Mariam↗

Delivering Science on Day One

While benchmarking software is useful for testing the performance limits and stability of Argonne National Laboratory’s new Theta supercomputer, there is no substitute for running real applications to explore the system’s potential. The Argonne Leadership Computing Facility’s Theta Early Science Program, modeled after its highly successful code migration program for the Mira supercomputer, has one primary aim: to deliver science on day one. Here is a closer look at the type of science problems that will be getting early access to Theta, a next-generation machine being rolled out this year.

Williams, Timothy J.↗

Shrink or Substitute: Handling Process Failures in HPC Systems Using In-Situ Recovery

Efficient utilization of today's high-performance computing (HPC) systems with complex software and hardware components requires that the HPC applications are designed to tolerate process failures at runtime. With low mean-time-to-failure (MTTF) of current and future HPC systems, long running simulations on these systems requires capabilities for gracefully handling process failures by the applications themselves. In this paper, we explore the use of fault tolerance extensions to Message Passing Interface (MPI) called user-level failure mitigation (ULFM) for handling process failures without the need to discard the progress made by the application. We explore two alternative recovery strategies, which use ULFM along with application-driven in-memory checkpointing. In the first case, the application is recovered with only the surviving processes, and in the second case, spares are used to replace the failed processes, such that the original configuration of the application is restored. Our experimental results demonstrate that graceful degradation is a viable alternative for recovery in environments where spares may not be available.

Ashraf, Rizwan↗

Direct Modeling of Scintillator Thickness for Optimal Light Output and Spatial Resolution

It is common in x-ray radiography to use scintillators (e.g., BGO or LSO) to convert x-rays to visible light, which is then recorded by an imaging system. The response of the scintillator depends fundamentally on its thickness, with respect to both its visible light emittance and its spatial resolution. This is important for optimizing light output, signal to noise ratio, or optical response time. Given that it is often cost-prohibitive to procure a variety of scintillator samples and empirically test the performance, it is essential to be able to model and accurately simulate the performance of a scintillator with respect to thickness and other properties, and a direct way of doing this is using Monte Carlo-based radiation transport codes. Such simulations can be expensive in terms of computational time, and the codes are not easily obtained. In this work we first show such simulations, and demonstrate that there is a natural trade-off between light output of a scintillator and its spatial resolution. We then derive a first-principles model that accurately approximates the light output, using straightforward calculations that can be performed quickly with any basic computing software. We compare the results to those obtained from Monte Carlo simulations and show that our simplified model can be used to analyze the tradeoff between emittance and resolution nearly as well as using a full-scale radiation transport code.

Monte Carlo methods, X-ray imaging, scintillation ↗

Real-Time Hardware-in-The-Loop Implementation of Protection and Self-Healing of Microgrids

The design of a reliable protection scheme for microgrids often requires communication between protective devices and microgrid controllers. The authors have developed such a communication assisted scalable protection scheme with a self healing feature to protect a microgrid with 100% Inverter Based Resources (IBRs). The communication assisted scheme must be validated in real time with cyber physical co-simulation for successful demonstration. In this regard, the paper presents a co-simulation platform between a simulated power system model using RTDS and physical protective devices. The primary protection of the scheme is programmed in SEL 421-7 relay and backup protection is programmed in MATLAB on a generic computer acting as a microgrid controller. The IEC 61850 models are used to communicate between SEL-421-7 relay and RTDS, whereas TCP/IP communication connects the microgrid controller to RTDS. The paper’s focus is to demonstrate the co- simulation platform with communication links established using both protocols and validate the proposed scheme in real-time on the IEEE 123 node distribution feeder. The paper explains the configuration of the IEC 61850 and TCP/IP communications as the interface requires proper hardware and software setup. Furthermore, the real time performance indicates the Hardware In the Loop (HIL) framework as a competent testing environment for the developed protection scheme for microgrids.

42 ENGINEERING↗

Illinois Express Quantum Network (IEQNET): metropolitan-scale experimental quantum networking over deployed optical fiber

The Illinois Express Quantum Network (IEQNET) is a program to realize metro-scale quantum networking over deployed optical fiber using currently available technology. IEQNET consists of multiple sites that are geographically dispersed in the Chicago metropolitan area. Each site has one or more quantum nodes (Qnodes) representing the communication parties in a quantum network. Q-nodes generate or measure quantum signals such as entangled photons and communicate the results via standard, classical, means. The entangled photons in IEQNET nodes are generated at multiple wavelengths, and are selectively distributed to the desired users via optical switches. Here we describe the network architecture of IEQNET, including the Internet-inspired layered hierarchy that leverages software-defined-networking (SDN) technology to perform traditional wavelength routing and assignment between the Q-nodes. Specifically, SDN decouples the control and data planes, with the control plane being entirely classical. Issues associated with synchronization, calibration, network monitoring, and scheduling will be discussed. An important goal of IEQNET is demonstrating the extent to which the control plane can coexist with the data plane using the same fiber lines. This goal is furthered by the use of tunable narrow-band optical filtering at the receivers and, at least in some cases, a wide wavelength separation between the quantum and classical channels. We envision IEQNET to aid in developing robust and practical quantum networks by demonstrating metro-scale quantum communication tasks such as entanglement distribution and quantum-state teleportation.

Chung, Joaquin↗

SU-C-201-01: Investigation of the Effects of Scintillator Surface Treatment On Light Output Measurements with SiPM Detectors

Purpose: To examine the effect of different scintillator surface treatments (BGO crystals) on the fraction of scintillation photons that exit the crystal and reach the photodetector (SiPM). Methods: Positron Emission Tomography is based on the detection of light that exits scintillator crystals, after annihilation photons deposit energy inside these crystals. A considerable fraction of the scintillation light gets trapped or absorbed after going through multiple internal reflections on the interfaces surrounding the crystals. BGO scintillator crystals generate considerably less scintillation light than crystals made of LSO and its variants. Therefore, it is crucial that the small amount of light produced by BGO exits towards the light detector. The surface treatment of scintillator crystals is among the factors affecting the ability of scintillation light to reach the detectors. In this study, we analyze the effect of different crystal surface treatments on the fraction of scintillation light that is detected by the solid state photodetector (SiPM), once energy is deposited inside a BGO crystal. Simulations were performed by a Monte Carlo based software named GATE, and validated by measurements from individual BGO crystals coupled to Philips digital-SiPM sensor (DPC-3200). Results: The results showed an increment in light collection of about 4 percent when only the exit face of the BGO crystal, is unpolished; compared to when all the faces are polished. However, leaving several faces unpolished caused a reduction of at least 10 percent of light output when the interaction occurs as far from the exit face of the crystal as possible compared to when it occurs very close to the exit face. Conclusion: This work demonstrates the advantages on light collection from leaving unpolished the exit face of BGO crystals. The configuration with best light output will be used to obtain flood images from BGO crystal arrays coupled to SiPM sensors.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

SU-E-J-09: Image Quality Comparison and Dose Quantification for 2.5 MV

Purpose: To compare the image quality of the 2.5MV imaging beam (2.5X-IMB) to that of a 6MV beam and to quantify the imaging dose of a 2.5X-IMB for constancy as specified by AAPM TG-142 Methods: The image quality of the 2.5X-IMB was compared to the 6MV imaging beam using the SNC ImagePro MV-QA phantom and the Varian supplied Las Vegas phantom (LVP). High resolution (1280×1280×16, 2 frames at 1.5MU/frame) and low resolution (640×640×16, 2 frames at 0.75MU/frame) images were compared for each phantom. MV-QA phantom images were evaluated quantitatively, and the LVP images were evaluated qualitatively. The imaging dose for 2.5X-IMB was quantified using the procedure outlined in TG51. PTWCC13-31013 chambers were used to measure a percent depth dose (PDD) curve for the 2.5X-IMB. All the factors described in TG51 were calculated using the 2.5X-IMB and a PTW30013 farmer chamber. Results: A comparison between 2.5X-IMB and 6MV image quality was performed both visually and with DoseLab software. The optimal window and level were set for each image of the LVP by the user. Visual inspection showed greater contrast resolution with the 2.5MV beam, but no significant difference with the change in imaging resolution. DoseLab reported similar spatial resolutions between the two energies, but the contrast-to-noise ratio (CNR) was greater for 2.5MV. The PDDx(10cm) for a 10x10cm2 field was measured to be 51.5%. Although this PDD value is off the scale of Figure 4 in TG51, the trend of the curve corresponding to the PTW31003 (equivalent) chamber led to an approximate kQ value of 1.00. Conclusion: When compared to 6MV imaging, 2.5X-IMB results in a better CNR. At low resolution, the DoseLab results for the two energies are comparable, but visual analysis favors the 2.5X-IMB images. Imaging dose was quantified for the 2.5X-IMB after following the TG51 methodology with appropriate approximations.

60 APPLIED LIFE SCIENCES↗

SU-E-J-182: Reproducibility of Tumor Motion Probability Distribution Function in Stereotactic Body Radiation Therapy of Lung Using Real-Time Tumor-Tracking Radiotherapy System

Purpose: We aim to achieve new four-dimensional radiotherapy (4DRT) using the next generation real-time tumor-tracking (RTRT) system and flattening-filter-free techniques. To achieve new 4DRT, it is necessary to understand the respiratory motion of tumor. The purposes of this study were: 1.To develop the respiratory motion analysis tool using log files. 2.To evaluate the reproducibility of tumor motion probability distribution function (PDF) during stereotactic body RT (SBRT) of lung tumor. Methods: Seven patients having fiducial markers closely implanted to the lung tumor were enrolled in this study. The positions of fiducial markers were measured using the RTRT system (Mitsubishi Electronics Co., JP) and recorded as two types of log files during the course of SBRT. For each patients, tumor motion range and tumor motion PDFs in left-right (LR), anterior-posterior (AP) and superior-inferior (SI) directions were calculated using log files of all beams per fraction (PDFn). Fractional PDF reproducibility (Rn) was calculated as Kullback-Leibler (KL) divergence between PDF1 and PDFn of tumor motion. The mean of Rn (Rm) was calculated for each patient and correlated to the patient’s mean tumor motion range (Am). The change of Rm during the course of SBRT was also evluated. These analyses were performed using in-house developed software. Results: The Rm were 0.19 (0.07–0.30), 0.14 (0.07–0.32) and 0.16 (0.09–0.28) in LR, AP and SI directions, respectively. The Am were 5.11 mm (2.58–9.99 mm), 7.81 mm (2.87–15.57 mm) and 11.26 mm (3.80–21.27 mm) in LR, AP and SI directions, respectively. The PDF reproducibility decreased as the tumor motion range increased in AP and SI direction. That decreased slightly through the course of RT in SI direction. Conclusion: We developed the respiratory motion analysis tool for 4DRT using log files and quantified the range and reproducibility of respiratory motion for lung tumors.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

TU-G-BRA-05: Predicting Volume Change of the Tumor and Critical Structures Throughout Radiation Therapy by CT-CBCT Registration with Local Intensity Correction

Purpose: The purpose of this study is to develop an accurate and effective technique to predict and monitor volume changes of the tumor and organs at risk (OARs) from daily cone-beam CTs (CBCTs). Methods: While CBCT is typically used to minimize the patient setup error, its poor image quality impedes accurate monitoring of daily anatomical changes in radiotherapy. Reconstruction artifacts in CBCT often cause undesirable errors in registration-based contour propagation from the planning CT, a conventional way to estimate anatomical changes. To improve the registration and segmentation accuracy, we developed a new deformable image registration (DIR) that iteratively corrects CBCT intensities using slice-based histogram matching during the registration process. Three popular DIR algorithms (hierarchical B-spline, demons, optical flow) augmented by the intensity correction were implemented on a graphics processing unit for efficient computation, and their performances were evaluated on six head and neck (HN) cancer cases. Four trained scientists manually contoured nodal gross tumor volume (GTV) on the planning CT and every other fraction CBCTs for each case, to which the propagated GTV contours by DIR were compared. The performance was also compared with commercial software, VelocityAI (Varian Medical Systems Inc.). Results: Manual contouring showed significant variations, [-76, +141]% from the mean of all four sets of contours. The volume differences (mean±std in cc) between the average manual segmentation and four automatic segmentations are 3.70±2.30(B-spline), 1.25±1.78(demons), 0.93±1.14(optical flow), and 4.39±3.86 (VelocityAI). In comparison to the average volume of the manual segmentations, the proposed approach significantly reduced the estimation error by 9%(B-spline), 38%(demons), and 51%(optical flow) over the conventional mutual information based method (VelocityAI). Conclusion: The proposed CT-CBCT registration with local CBCT intensity correction can accurately predict the tumor volume change with reduced errors. Although demonstrated only on HN nodal GTVs, the results imply improved accuracy for other critical structures. This work was supported by NIH/NCI under grant R42CA137886.

60 APPLIED LIFE SCIENCES↗

SU-F-T-185: Study of the Robustness of a Proton Arc Technique Based On PBS Beams

Purpose: One potential technique to realize proton arc is through using PBS beams from many directions to form overlaid Bragg peak (OBP) spots and placing these OBP spots throughout the target volume to achieve desired dose distribution. In this study, we analyzed the robustness of this proton arc technique. Methods: We used a cylindrical water phantom of 20 cm in radius in our robustness analysis. To study the range uncertainty effect, we changed the density of the phantom by ±3%. To study the setup uncertainty effect, we shifted the phantom by 3 & 5 mm. We also combined the range and setup uncertainties (3mm/±3%). For each test plan, we performed dose calculation for the nominal and 6 disturbed scenarios. Two test plans were used, one with single OBP spot and the other consisting of 121 OBP spots covering a 10×10cm{sup 2} area. We compared the dose profiles between the nominal and disturbed scenarios to estimate the impact of the uncertainties. Dose calculation was performed with Gate/GEANT based Monte Carlo software in cloud computing environment. Results: For each of the 7 scenarios, we simulated 100k & 10M events for plans consisting of single OBP spot and 121 OBP spots respectively. For single OBP spot, the setup uncertainty had minimum impact on the spot’s dose profile while range uncertainty had significant impact on the dose profile. For plan consisting of 121 OBP spots, similar effect was observed but the extent of disturbance was much less compared to single OBP spot. Conclusion: For PBS arc technique, range uncertainty has significantly more impact than setup uncertainty. Although single OBP spot can be severely disturbed by the range uncertainty, the overall effect is much less when a large number of OBP spots are used. Robustness optimization for PBS arc technique should consider range uncertainty with priority.

60 APPLIED LIFE SCIENCES↗

Impact of compressed breast thickness and dose on lesion detectability in digital mammography: FROC study with simulated lesions in real mammograms

Purpose: The aim of this work was twofold: (1) to examine whether, with standard automatic exposure control (AEC) settings that maintain pixel values in the detector constant, lesion detectability in clinical images decreases as a function of breast thickness and (2) to verify whether a new AEC setup can increase lesion detectability at larger breast thicknesses. Methods: Screening patient images, acquired on two identical digital mammography systems, were collected over a period of 2 yr. Mammograms were acquired under standard AEC conditions (part 1) and subsequently with a new AEC setup (part 2), programmed to use the standard AEC settings for compressed breast thicknesses ≤49 mm, while a relative dose increase was applied above this thickness. The images were divided into four thickness groups: T1 ≤ 29 mm, T2 = 30–49 mm, T3 = 50–69 mm, and T4 ≥ 70 mm, with each thickness group containing 130 randomly selected craniocaudal lesion-free images. Two measures of density were obtained for every image: a BI-RADS score and a map of volumetric breast density created with a software application (VolparaDensity, Matakina, NZ). This information was used to select subsets of four images, containing one image from each thickness group, matched to a (global) BI-RADS score and containing a region with the same (local) VOLPARA volumetric density value. One selected lesion (a microcalcification cluster or a mass) was simulated into each of the four images. This process was repeated so that, for a given thickness group, half the images contained a single lesion and half were lesion-free. The lesion templates created and inserted in groups T3 and T4 for the first part of the study were then inserted into the images of thickness groups T3 and T4 acquired with higher dose settings. Finally, all images were visualized using the ViewDEX software and scored by four radiologists performing a free search study. A statistical jackknife-alternative free-response receiver operating characteristic analysis was applied. Results: For part 1, the alternative free-response receiver operating characteristic curves for the four readers were 0.80, 0.65, 0.55 and 0.56 in going from T1 to T4, indicating a decrease in detectability with increasing breast thickness. P-values and the 95% confidence interval showed no significant difference for the T3-T4 comparison (p = 0.78) while all the other differences were significant (p < 0.05). Separate analysis of microcalcification clusters presented the same results while for mass detection, the only significant difference came when comparing T1 to the other thickness groups. Comparing the scores of part 1 and part 2, results for the T3 group acquired with the new AEC setup and T3 group at standard AEC doses were significantly different (p = 0.0004), indicating improved detection. For this group a subanalysis for microcalcification detection gave the same results while no significant difference was found for mass detection. Conclusions: These data using clinical images confirm results found in simple QA tests for many mammography systems that detectability falls as breast thickness increases. Results obtained with the AEC setup for constant detectability above 49 mm showed an increase in lesion detection with compressed breast thickness, bringing detectability of lesions to the same level.

60 APPLIED LIFE SCIENCES↗

A Distributed-Memory Package for Dense Hierarchically Semi-Separable Matrix Computations Using Randomization

In this paper, we present a distributed-memory library for computations with dense structured matrices. A matrix is considered structured if its off-diagonal blocks can be approximated by a rank-deficient matrix with low numerical rank. Here, we use Hierarchically Semi-Separable (HSS) representations. Such matrices appear in many applications, for example, finite-element methods, boundary element methods, and so on. Exploiting this structure allows for fast solution of linear systems and/or fast computation of matrix-vector products, which are the two main building blocks of matrix computations. The compression algorithm that we use, that computes the HSS form of an input dense matrix, relies on randomized sampling with a novel adaptive sampling mechanism. We discuss the parallelization of this algorithm and also present the parallelization of structured matrix-vector product, structured factorization, and solution routines. The efficiency of the approach is demonstrated on large problems from different academic and industrial applications, on up to 8,000 cores. Finally, this work is part of a more global effort, the STRUctured Matrices PACKage (STRUMPACK) software package for computations with sparse and dense structured matrices. Hence, although useful on their own right, the routines also represent a step in the direction of a distributed-memory sparse solver.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗