Improving Chemical Mapping Algorithm and Visualization in Full-field Hard X-ray Spectroscopic Imaging
N/A
97 MATHEMATICS AND COMPUTING↗
SEARCH · Engineering Papers
Results for “parallel algorithms”
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.
Scalability of high-performance PDE solvers
Performance tests and analyses are critical to effective high-performance computing software development and are central components in the design and implementation of computational algorithms for achieving faster simulations on existing and future computing architectures for large-scale application problems. In this article, we explore performance and space-time trade-offs for important compute-intensive kernels of large-scale numerical solvers for partial differential equations (PDEs) that govern a wide range of physical applications. We consider a sequence of PDE-motivated bake-off problems designed to establish best practices for efficient high-order simulations across a variety of codes and platforms. We measure peak performance (degrees of freedom per second) on a fixed number of nodes and identify effective code optimization strategies for each architecture. In addition to peak performance, we identify the minimum time to solution at 80% parallel efficiency. The performance analysis is based on spectral and p-type finite elements but is equally applicable to a broad spectrum of numerical PDE discretizations, including finite difference, finite volume, and h-type finite elements.
97 MATHEMATICS AND COMPUTING↗
Machine Learning Based Parallel I/O Predictive Modeling: A Case Study on Lustre File Systems
Parallel I/O hardware and software infrastructure is a key contributor to performance variability for applications running on large-scale HPC systems. This variability confounds efforts to predict application performance for characterization, modeling, optimization, and job scheduling. We propose a modeling approach that improves predictive ability by explicitly treating the variability and by leveraging the sensitivity of application parameters on performance to group applications with similar characteristics. We develop a Gaussian process-based machine learning algorithm to model I/O performance and its variability as a function of application and file system characteristics. We demonstrate the effectiveness of the proposed approach using data collected from the Edison system at the National Energy Research Scientific Computing Center. The results show that the proposed sensitivity-based models are better at prediction when compared with application-partitioned or unpartitioned models. We highlight modeling techniques that are robust to the outliers that can occur in production parallel file systems. Using the developed metrics and modeling approach, we provide insights into the file system metrics that have a significant impact on I/O performance.
I/O performance variability↗
CRPropa 3.1—a low energy extension based on stochastic differential equations
The propagation of charged cosmic rays through the Galactic environment influences all aspects of the observation at Earth. Energy spectrum, composition and arrival directions are changed due to deflections in magnetic fields and interactions with the interstellar medium. Today the transport is simulated with different simulation methods either based on the solution of a transport equation (multi-particle picture) or a solution of an equation of motion (single-particle picture). We developed a new module for the publicly available propagation software CRPropa 3.1, where we implemented an algorithm to solve the transport equation using stochastic differential equations. This technique allows us to use a diffusion tensor which is anisotropic with respect to an arbitrary magnetic background field. The source code of CRPropa is written in C++ with python steering via SWIG which makes it easy to use and computationally fast. In this paper, we present the new low-energy propagation code together with validation procedures that are developed to proof the accuracy of the new implementation. Furthermore, we show first examples of the cosmic ray density evolution, which depends strongly on the ratio of the parallel κ{sub ∥} and perpendicular κ{sub ⊥} diffusion coefficients. This dependency is systematically examined as well the influence of the particle rigidity on the diffusion process.
79 ASTRONOMY AND ASTROPHYSICS↗
A massively parallel time-domain coupled electrodynamics–micromagnetics solver
We present a high-performance coupled electrodynamics–micromagnetics solver for full physical modeling of signals in microelectronic circuitry. The overall strategy couples a finite-difference time-domain approach for Maxwell’s equations to a magnetization model described by the Landau–Lifshitz–Gilbert equation. The algorithm is implemented in the Exascale Computing Project software framework, AMReX, which provides effective scalability on manycore and GPU-based supercomputing architectures. Furthermore, the code leverages ongoing developments of the Exascale Application Code, WarpX, which is primarily being developed for plasma wakefield accelerator modeling. Our temporal coupling scheme provides second-order accuracy in space and time by combining the integration steps for the magnetic field and magnetization into an iterative sub-step that includes a trapezoidal temporal discretization for the magnetization. The performance of the algorithm is demonstrated by the excellent scaling results on NERSC multicore and GPU systems, with a significant (59×) speedup on the GPU using a node-by-node comparison. We demonstrate the utility of our code by performing simulations of an electromagnetic waveguide and a magnetically tunable filter.
97 MATHEMATICS AND COMPUTING↗
A new correction method serving to eliminate the parabola effect of flatbed scanners used in radiochromic film dosimetry
Purpose: The purpose of this study is the correction of the lateral scanner artifact, i.e., the effect that, on a large homogeneously exposed EBT3 film, a flatbed scanner measures different optical densities at different positions along thex axis, the axis parallel to the elongated light source. At constant dose, the measured optical densitiy profiles along this axis have a parabolic shape with significant dose dependent curvature. Therefore, the effect is shortly called the parabola effect. The objective of the algorithm developed in this study is to correct for the parabola effect. Any optical density measured at given position x is transformed into the equivalent optical density c at the apex of the parabola and then converted into the corresponding dose via the calibration of c versus dose. Methods: For the present study EBT3 films and an Epson 10000XL scanner including transparency unit were used for the analysis of the parabola effect. The films were irradiated with 6 MV photons from an Elekta Synergy accelerator in a RW3 slab phantom. In order to quantify the effect, ten film pieces with doses graded from 0 to 20.9 Gy were sequentially scanned at eight positions along thex axis and at six positions along the z axis (the movement direction of the light source) both for the portrait and landscape film orientations. In order to test the effectiveness of the new correction algorithm, the dose profiles of an open square field and an IMRT plan were measured by EBT3 films and compared with ionization chamber and ionization chamber array measurement. Results: The parabola effect has been numerically studied over the whole measuring field of the Epson 10000XL scanner for doses up to 20.9 Gy and for both film orientations. The presented algorithm transforms any optical density at positionx into the equivalent optical density that would be measured at the same dose at the apex of the parabola. This correction method has been validated up to doses of 5.2 Gy all over the scanner bed with 2D dose distributions of an open square photon field and an IMRT distribution. Conclusions: The algorithm presented in this study quantifies and corrects the parabola effect of EBT3 films scanned in commonly used commercial flatbed scanners at doses up to 5.2 Gy. It is easy to implement, and no additional work steps are necessary in daily routine film dosimetry.
61 RADIATION PROTECTION AND DOSIMETRY↗
SU-F-T-151: Measurement Evaluation of Skin Dose in Scanning Proton Beam Therapy for Breast Cancer
Purpose: To measure the skin dose and compare it with the calculated dose from a treatment planning system (TPS) for breast cancer treatment using scanning proton beam therapy (SPBT). Methods: A single en-face-beam SPBT plan was generated by a commercial TPS for two breast cancer patients. The treatment volumes were the entire breasts (218 cc and 1500 cc) prescribed to 50.4 Gy (RBE) in 28 fractions. A range shifter of 5 cm water equivalent thickness was used. The organ at risk (skin) was defined to be 5 mm thick from the surface. The skin doses were measured in water with an ADCL calibrated parallel plate (PP) chamber. The measured data were compared with the values calculated in the TPS. Skin dose calculations can be subject to uncertainties created by the definition of the external contour and the limitations of the correction based algorithms, such as proton convolution superposition. Hence, the external contours were expanded by 0, 3 mm and 1 cm to include additional pixels for dose calculation. In addition, to examine the effects of the cloth gown on the skin dose, the skin dose measurements were conducted with and without gown. Results: On average the measured skin dose was 4% higher than the calculated values. At deeper depths, the measured and calculated doses were in better agreement (< 2%). Large discrepancy occur for the dose calculated without external expansion due to volume averaging. The addition of the gown only increased the measured skin dose by 0.4%. Conclusion: The implemented TPS underestimated the skin dose for breast treatments. Superficial dose calculation without external expansion would result in large errors for SPBT for breast cancer.
60 APPLIED LIFE SCIENCES↗
Oineus v1.0
A library for multi-threaded computation of persistence diagrams. The algorithm for computing persistence diagrams in a lock-free manner was published in the 'Towards Lock-free Persistent Homology' (D. Morozov, A. Nigmetov, Brief Announcement: SPAA 2020); it scales better than the only other shared-memory parallel implementation of persistent homology computation PHAT. Library includes python bindings to compute persistence diagrams and their vectorizations for lower-star filtrations on grid data. It is intended to be used by scientists working on Topological Data Analysis and its applications in different areas.
Nigmetov, Arnur↗
mystic - a brief introduction
The mystic framework enables the solving of difficult optimization problems through three major innovations: (1) plug-and-play component-based optimization algorithms, (2) the ability to dynamically construct and apply generalzed trnasforms, K , where x' = K(x) , and (3) workflows providing asynchronous execution and statefulness. These features, in conjunction with supporting packages pathos , pyina , and klepto , can provide massively-parallel scalable workflows for quickly solving optimization problems in complex nonlinear spaces. mystic can be used to simplify non-convex optimization problems by transforming away nonlinearities through user-built kernel transforms. This tutorial demonstrates using mystic for machine learning with automated dimensional-reduction, using embarrasingly parallel solver ensembles to find an accurate interpolated surrogate for a nonlinear surface, and the determination of worst-case bounds on expectation value of an objective function under uncertainty.
97 MATHEMATICS AND COMPUTING↗
Algorithm 1028: VTMOP: Solver for Blackbox Multiobjective Optimization Problems
VTMOP is a Fortran 2008 software package containing two Fortran modules for solving computationally expensive bound-constrained blackbox multiobjective optimization problems. VTMOP implements the algorithm of [32], which handles two or more objectives, does not require any derivatives, and produces well-distributed points over the Pareto front. The first module contains a general framework for solving multiobjective optimization problems by combining response surface methodology, trust region methodology, and an adaptive weighting scheme. The second module features a driver subroutine that implements this framework when the objective functions can be wrapped as a Fortran subroutine. Lastly, support is provided for both serial and parallel execution paradigms, and VTMOP is demonstrated on several test problems as well as one real-world problem in the area of particle accelerator optimization.
97 MATHEMATICS AND COMPUTING↗
Towards Superior Software Portability with SHAD and HPX C++ Libraries
As hardware architectures and software stacks complexity grows, development productivity, performance and software portability, quickly evolve from desirable features to actual needs. SHAD, the Scalable High-performance Algorithms and Data-structures C++ library is designed to mitigate these issues: it provides general purpose building blocks as well as high-level custom utilities, and offers a shared-memory programming abstraction which facilitates the programming of complex systems, scaling up to High Performance Computing clusters. SHAD’s portability is achieved through an abstract runtime interface, which decouples the upper layers of the library and hides the low level details of the underlying architecture. This layer enables SHAD to interface with different runtime/threading systems, e.g. Intel TBB and Global Memory and Threading (GMT). However, current backends targeting distributed systems, rely on a centralized controller which may possibly limit scalability up to hundreds of nodes and creates a network hot spot due to all to one communication for synchronization, and possibly resulting in degraded performance at high process counts. In this research, we explore HPX, the C++ standard library for parallelism and concurrency, as an additional backend in support of the SHAD library, and present the methodologies in support of local and remote task executions in SHAD with respect to HPX. Finally, we evaluate the proposed system by comparing against existing backends of SHAD and analyzing their performance on C++ Standard Template Library algorithms.
Wu, Nanmiao↗
SNoGloDe: A Structured Nonlinear Global Decomposition Solver
Large-scale optimization problems often require decomposition strategies and customized algorithms to achieve optimal solutions within a reasonable time. Building on the work of Cao and Zavala (2019) for solving nonlinear two-stage stochastic programs to global optimality, we implement and extend their approach. We generalize to optimization problems reformulated with a block-angular constraint structure (e.g., temporal decomposition). Our framework, written in Python using Pyomo, is highly customizable and enables parallel execution of the decomposition. SNoGloDe allows tailored branching strategies, lower bounding problems, and candidate generators to leverage problem-specific knowledge. To demonstrate effectiveness, we compare SNoGloDe’s performance with Gurobi on a temporally decomposed produced water case study.
algorithms↗
SU-E-I-36: A KWIC and Dirty Look at Dose Savings and Perfusion Metrics in Simulated CT Neuro Perfusion Exams
Purpose: CT neuro perfusion scans are one of the highest dose exams. Methods to reduce dose include decreasing the number of projections acquired per gantry rotation, however conventional reconstruction of such scans leads to sampling artifacts. In this study we investigated a projection view-sharing reconstruction algorithm used in dynamic MRI – “K-space Weighted Image Contrast” (KWIC) – applied to simulated perfusion exams and evaluated dose savings and impacts on perfusion metrics. Methods: A FORBILD head phantom containing simulated time-varying objects was developed and a set of parallel-beam CT projection data was created. The simulated scans were 60 seconds long, 1152 projections per turn, with a rotation time of one second. No noise was simulated. 5mm, 10mm, and 50mm objects were modeled in the brain. A baseline, “full dose” simulation used all projections and reduced dose cases were simulated by downsampling the number of projections per turn from 1152 to 576 (50% dose), 288 (25% dose), and 144 (12.5% dose). KWIC was further evaluated at 72 projections per rotation (6.25%). One image per second was reconstructed using filtered backprojection (FBP) and KWIC. KWIC reconstructions utilized view cores of 36, 72, 144, and 288 views and 16, 8, 4, and 2 subapertures respectively. From the reconstructed images, time-to-peak (TTP), cerebral blood flow (CBF) and the FWHM of the perfusion curve were calculated and compared against reference values from the full-dose FBP data. Results: TTP, CBF, and the FWHM were unaffected by dose reduction (to 12.5%) and reconstruction method, however image quality was improved when using KWIC. Conclusion: This pilot study suggests that KWIC preserves image quality and perfusion metrics when under-sampling projections and that the unique contrast weighting of KWIC could provided substantial dose-savings for perfusion CT scans. Evaluation of KWIC in clinical CT data will be performed in the near future. R01 EB014922, NCI Grant U01 CA181156 (Quantitative Imaging Network), and Tobacco Related Disease Research Project grant 22RT-0131.
60 APPLIED LIFE SCIENCES↗
Addressing Load Imbalance in Bioinformatics and Biomedical Applications: Efficient Scheduling across Multiple GPUs
Computational bioinformatics and biomedical applications frequently contain heterogeneously sized units of work or tasks, for instance due to variability in the sizes of biological sequences and molecules. Variable-sized workloads lead to load imbalances in parallel implementations which detract from efficiency and performance. Many modern computing resources now have multiple graphics processing units(GPUs) per computer for acceleration. These multiple GPU resources need to be used efficiently through balancing of workloads across the GPUs. OpenMP is a portable directive-based parallel programming API used ubiquitously in bioscience applications to program CPUs; recently, the use of OpenMP directives for GPU acceleration has become possible. Here, motivated by experiences with imbalanced loads in GPU-accelerated bioinformatics applications, we address the load balancing problem using OpenMP task-to-GPU scheduling combined with OpenMP GPU offloading for multiply heterogeneous workloads – loads with both variable input sizes, and simultaneously, variable convergence rates for algorithms with a stochastic component – scheduled across multiple GPUs. We aim to develop strategies which are both easy to use and have lower overheads, and may be incorporated incrementally in existing programs which already make use of OpenMP for CPU-based threading in order to make use of multi-GPU computers. We test different combinations of input size variability and convergence rate variability, and characterize the effects of these different scenarios on the performance of scheduling strategies across multiple GPUs with OpenMP. We present several dynamic scheduling solutions for different parallel patterns, explore optimizations, and provide publicly available example computational kernels to make these strategies easy to use in programs. This work will enable application developers to efficiently and easily use multiple GPUs for imbalanced workloads found in bioinformatics and biomedical applications.
Thavappiragasam, Mathialakan↗
molgw 1: Many-body perturbation theory software for atoms, molecules, and clusters
Here, we summarize the MOLGW code that implements density-functional theory and many-body perturbation theory in a Gaussian basis set. The code is dedicated to the calculation of the many-body self-energy within the GW approximation and the solution of the Bethe–Salpeter equation. These two types of calculations allow the user to evaluate physical quantities that can be compared to spectroscopic experiments. Quasiparticle energies, obtained through the calculation of the GW self-energy, can be compared to photoemission or transport experiments, and neutral excitation energies and oscillator strengths, obtained via solution of the Bethe–Salpeter equation, are measurable by optical absorption. The implementation choices outlined here have aimed at the accuracy and robustness of calculated quantities with respect to measurements. Furthermore, the algorithms implemented in MOLGW allow users to consider molecules or clusters containing up to 100 atoms with rather accurate basis sets, and to choose whether or not to apply the resolution-of-the-identity approximation. Finally, we demonstrate the parallelization efficacy of the MOLGW code over several hundreds of processors.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
Did the GPU obfuscate the load imbalance in my MPI simulation?
The current proliferation of GPU-based HPC systems necessitates a method for assessing the performance of simulations on heterogeneous machines. The addition of GPUs to a system adds multiple hierarchical levels of parallelism to the node architecture. In this paper, we demonstrate that the traditional load imbalance metric is insufficient for capturing the load imbalance on GPU-based machines, since it treats the GPU as a monolithic entity and ignores the internal parallelism. We propose a new hierarchical metric that improves the correlation of measured performance and application workload by up to 20.61%. Using our metric for determining application load instead of the traditional metric as the input for the load balancing algorithm reduces the residual load imbalance by up to 4× in our application.
Eberius, David↗
Scalable Deep Learning via I/O Analysis and Optimization
Scalable deep neural network training has been gaining prominence because of the increasing importance of deep learning in a multitude of scientific and commercial domains. Consequently, a number of researchers have investigated techniques to optimize deep learning systems. Much of the prior work has focused on runtime and algorithmic enhancements to optimize the computation and communication. Despite these enhancements, however, deep learning systems still suffer from scalability limitations, particularly with respect to data I/O. This situation is especially true for training models where the computation can be effectively parallelized, leaving I/O as the major bottleneck. In fact, our analysis shows that I/O can take up to 90% of the total training time. Thus, in this article, we first analyze LMDB, the most widely used I/O subsystem of deep learning frameworks, to understand the causes of this I/O inefficiency. Based on our analysis, we propose LMDBIO—an optimized I/O plugin for scalable deep learning. LMDBIO includes six novel optimizations that together address the various shortcomings in existing I/O for deep learning. Lastly, our experimental results show that LMDBIO significantly outperforms LMDB in all cases and improves overall application performance by up to 65-fold on a 9,216-core system.
97 MATHEMATICS AND COMPUTING↗
Global Pathway Selection with Zero-RK
Global Pathway Selection (GPS) is an algorithm to effectively generates reduced (skeletal) chemistry mechanisms, which speeds up simulations and can be used as a systematic analytics tool to extract insights from complex reacting system. This release is an extension of the original code to run in parallel and to use LLNL's Zero-RK solver for fast solution of chemical problems.
Whitesides, RussellA↗