Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Sparse matrix”

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 145 records · Page 8

High-order commutator-free exponential time-propagation of driven quantum systems

We discuss the numerical solution of the Schroedinger equation with a time-dependent Hamilton operator using commutator-free time-propagators. These propagators are constructed as products of exponentials of simple weighted sums of the Hamilton operator. Owing to their exponential form they strictly preserve the unitarity of time-propagation. The absence of commutators or other computationally involved operations allows for straightforward implementation and application also to large scale and sparse matrix problems. We explain the derivation of commutator-free exponential time-propagators in the context of the Magnus expansion, and provide optimized propagators up to order eight. An extensive theoretical error analysis is presented together with practical efficiency tests for different problems. Issues of practical implementation, in particular the use of the Krylov technique for the calculation of exponentials, are discussed. We demonstrate for two advanced examples, the hydrogen atom in an electric field and pumped systems of multiple interacting two-level systems or spins that this approach enables fast and accurate computations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Basker: Parallel sparse LU factorization utilizing hierarchical parallelism and data layouts

Transient simulation in circuit simulation tools, such as SPICE and Xyce, depend on scalable and robust sparse LU factorizations for efficient numerical simulation of circuits and power grids. As the need for simulations of very large circuits grow, the prevalence of multicore architectures enable us to use shared memory parallel algorithms for such simulations. A parallel factorization is a critical component of such shared memory parallel simulations. We develop a parallel sparse factorization algorithm that can solve problems from circuit simulations efficiently, and map well to architectural features. This new factorization algorithm exposes hierarchical parallelism to accommodate irregular structure that arise in our target problems. It also uses a hierarchical two-dimensional data layout which reduces synchronization costs and maps to memory hierarchy found in multicore processors. We present an OpenMP based implementation of the parallel algorithm in a new multithreaded solver called Basker in the Trilinos framework. Here, we present performance evaluations of Basker on the Intel SandyBridge and Xeon Phi platforms using circuit and power grid matrices taken from the University of Florida sparse matrix collection and from Xyce circuit simulation. Basker achieves a geometric mean speedup of 5.91× on CPU (16 cores) and 7.4× on Xeon Phi (32 cores) relative to state-of-the-art solver KLU. Basker outperforms Intel MKL Pardiso solver (PMKL) by as much as 30× on CPU (16 cores) and 7.5× on Xeon Phi (32 cores) for low fill-in circuit matrices. Furthermore, Basker provides 5.4× speedup on a challenging matrix sequence taken from an actual Xyce simulation.

97 MATHEMATICS AND COMPUTING↗

BLINK enables ultrafast tandem mass spectrometry cosine similarity scoring

Metabolomics has a long history of using cosine similarity to match experimental tandem mass spectra to databases for compound identification. Here we introduce the Blur-and-Link (BLINK) approach for scoring cosine similarity. By bypassing fragment alignment and simultaneously scoring all pairs of spectra using sparse matrix operations, BLINK is over 3000 times faster than MatchMS, a widely used loop-based alignment and scoring implementation. Using a similarity cutoff of 0.7, BLINK and MatchMS had practically equivalent identification agreement, and greater than 99% of their scores and matching ion counts were identical. This performance improvement can enable calculations to be performed that would typically be limited by time and available computational resources.

47 OTHER INSTRUMENTATION↗

Solution of the Two-Dimensional Time-Dependent Schroedinger Equation Applied to Nuclear Proton Decay

A rigorous approach to study the temporal evolution of physical processes is to follow the development in time of a given initial state, by numerically solving the time-dependent Schroedinger equation. This represents a natural modeling of the dynamical behaviour. We considered the equation in two spatial coordinates, to describe deformed nuclear shapes. The Hamiltonian is discretized by special, functionally fitted difference formulae of the derivatives and then a Crank-Nicolson scheme is applied. The resulting linear system with large sparse matrix is solved by a variant of Conjugate Gradient Method. The numerical solution has been used to the description of the proton decay. We also discuss the treatment of numerical boundary conditions, the preparation of the initial wavefunction and the calculation of the decay rate through the flux.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Daubechies wavelets for linear scaling density functional theory

We demonstrate that Daubechies wavelets can be used to construct a minimal set of optimized localized adaptively contracted basis functions in which the Kohn-Sham orbitals can be represented with an arbitrarily high, controllable precision. Ground state energies and the forces acting on the ions can be calculated in this basis with the same accuracy as if they were calculated directly in a Daubechies wavelets basis, provided that the amplitude of these adaptively contracted basis functions is sufficiently small on the surface of the localization region, which is guaranteed by the optimization procedure described in this work. This approach reduces the computational costs of density functional theory calculations, and can be combined with sparse matrix algebra to obtain linear scaling with respect to the number of electrons in the system. Calculations on systems of 10 000 atoms or more thus become feasible in a systematic basis set with moderate computational resources. Further computational savings can be achieved by exploiting the similarity of the adaptively contracted basis functions for closely related environments, e.g., in geometry optimizations or combined calculations of neutral and charged systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Cholesky-decomposed density MP2 with density fitting: Accurate MP2 and double-hybrid DFT energies for large systems

Our recently developed QQR-type integral screening is introduced in our Cholesky-decomposed pseudo-densities Møller-Plesset perturbation theory of second order (CDD-MP2) method. We use the resolution-of-the-identity (RI) approximation in combination with efficient integral transformations employing sparse matrix multiplications. The RI-CDD-MP2 method shows an asymptotic cubic scaling behavior with system size and a small prefactor that results in an early crossover to conventional methods for both small and large basis sets. We also explore the use of local fitting approximations which allow to further reduce the scaling behavior for very large systems. The reliability of our method is demonstrated on test sets for interaction and reaction energies of medium sized systems and on a diverse selection from our own benchmark set for total energies of larger systems. Timings on DNA systems show that fast calculations for systems with more than 500 atoms are feasible using a single processor core. Parallelization extends the range of accessible system sizes on one computing node with multiple cores to more than 1000 atoms in a double-zeta basis and more than 500 atoms in a triple-zeta basis.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Modeling optical properties of silicon clusters by first principles: From a few atoms to large nanocrystals

Time dependent density functional tight binding (TDDFTB) method is implemented with sparse matrix techniques and improved parallelization algorithms. The method is employed to calculate the optical properties of various Si nanocrystals (NCs). The calculated light absorption spectra of small Si NCs from TDDFTB were found to be comparable with many body perturbation methods utilizing planewave basis sets. For large Si NCs (more than a thousand atoms) that are beyond the reach of conventional approaches, the TDDFTB method is able to produce reasonable results that are consistent with prior experiments. We also employed the method to study the effects of surface chemistry on the optical properties of large Si NCs. We learned that the optical properties of Si NCs can be manipulated with small molecule passivations such as methyl, hydroxyl, amino, and fluorine. In general, the shifts and profiles in the absorption spectra can be tuned with suitably chosen passivants.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Real-space solution to the electronic structure problem for nearly a million electrons

We report a Kohn–Sham density functional theory calculation of a system with more than 200 000 atoms and 800 000 electrons using a real-space high-order finite-difference method to investigate the electronic structure of large spherical silicon nanoclusters. Our system of choice was a 20 nm large spherical nanocluster with 202 617 silicon atoms and 13 836 hydrogen atoms used to passivate the dangling surface bonds. To speed up the convergence of the eigenspace, we utilized Chebyshev-filtered subspace iteration, and for sparse matrix–vector multiplications, we used blockwise Hilbert space-filling curves, implemented in the PARSEC code. For this calculation, we also replaced our orthonormalization + Rayleigh–Ritz step with a generalized eigenvalue problem step. We utilized all of the 8192 nodes (458 752 processors) on the Frontera machine at the Texas Advanced Computing Center. We achieved two Chebyshev-filtered subspace iterations, yielding a good approximation of the electronic density of states. Our work pushes the limits on the capabilities of the current electronic structure solvers to nearly 106 electrons and demonstrates the potential of the real-space approach to efficiently parallelize large calculations on modern high-performance computing platforms.

Chemistry↗

MADmap: A MASSIVELY PARALLEL MAXIMUM LIKELIHOOD COSMIC MICROWAVE BACKGROUND MAP-MAKER

MADmap is a software application used to produce maximum likelihood images of the sky from time-ordered data which include correlated noise, such as those gathered by cosmic microwave background (CMB) experiments. It works efficiently on platforms ranging from small workstations to the most massively parallel supercomputers. Map-making is a critical step in the analysis of all CMB data sets, and the maximum likelihood approach is the most accurate and widely applicable algorithm; however, it is a computationally challenging task. This challenge will only increase with the next generation of ground-based, balloon-borne, and satellite CMB polarization experiments. The faintness of the B-mode signal that these experiments seek to measure requires them to gather enormous data sets. MADmap is already being run on up to O(10{sup 11}) time samples, O(10{sup 8}) pixels, and O(10{sup 4}) cores, with ongoing work to scale to the next generation of data sets and supercomputers. We describe MADmap's algorithm based around a preconditioned conjugate gradient solver, fast Fourier transforms, and sparse matrix operations. We highlight MADmap's ability to address problems typically encountered in the analysis of realistic CMB data sets and describe its application to simulations of the Planck and EBEX experiments. The massively parallel and distributed implementation is detailed and scaling complexities are given for the resources required. MADmap is capable of analyzing the largest data sets now being collected on computing resources currently available, and we argue that, given Moore's Law, MADmap will be capable of reducing the most massive projected data sets.

79 ASTRONOMY AND ASTROPHYSICS↗

Algorithm for simulation of quantum many-body dynamics using dynamical coarse-graining

An algorithm for simulation of quantum many-body dynamics having su(2) spectrum-generating algebra is developed. The algorithm is based on the idea of dynamical coarse-graining. The original unitary dynamics of the target observables--the elements of the spectrum-generating algebra--is simulated by a surrogate open-system dynamics, which can be interpreted as weak measurement of the target observables, performed on the evolving system. The open-system state can be represented by a mixture of pure states, localized in the phase space. The localization reduces the scaling of the computational resources with the Hilbert-space dimension n by factor n{sup 3/2}(ln n){sup -1} compared to conventional sparse-matrix methods. The guidelines for the choice of parameters for the simulation are presented and the scaling of the computational resources with the Hilbert-space dimension of the system is estimated. The algorithm is applied to the simulation of the dynamics of systems of 2x10{sup 4} and 2x10{sup 6} cold atoms in a double-well trap, described by the two-site Bose-Hubbard model.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

The plug-based nanovolume Microcapillary Protein Crystallization System (MPCS)

The Microcapillary Protein Crystallization System (MPCS) embodies a new semi-automated plug-based crystallization technology which enables nanolitre-volume screening of crystallization conditions in a plasticware format that allows crystals to be easily removed for traditional cryoprotection and X-ray diffraction data collection. Protein crystals grown in these plastic devices can be directly subjected to in situ X-ray diffraction studies. The MPCS integrates the formulation of crystallization cocktails with the preparation of the crystallization experiments. Within microfluidic Teflon tubing or the microfluidic circuitry of a plastic CrystalCard, {approx}10-20 nl volume droplets are generated, each representing a microbatch-style crystallization experiment with a different chemical composition. The entire protein sample is utilized in crystallization experiments. Sparse-matrix screening and chemical gradient screening can be combined in one comprehensive 'hybrid' crystallization trial. The technology lends itself well to optimization by high-granularity gradient screening using optimization reagents such as precipitation agents, ligands or cryoprotectants.

59 BASIC BIOLOGICAL SCIENCES↗

Detector alignment for X-ray crystallography using Millepede-II

I describe a method for accurately refining the geometrical parameters of segmented X-ray area detectors on the basis of serial crystallography data, using 'Millepede' – an algorithm created for a very similar problem in high-energy physics. The Millepede method for serial crystallography builds on the approach of Brewster et al. [Acta Cryst. (2018), D74, 877–894], in which the detector parameters are refined simultaneously with the parameters for each individual crystal. This accounts for the mutual dependency between the parameters and thereby avoids the bias and slow convergence problems that have afflicted older approaches in which the deviations between observed and calculated Bragg peak positions were taken directly as the updates for the detector panel positions. The Millepede method uses the special structure of the least-squares normal equations to reduce them to a much smaller form that can be solved very quickly, even compared with the sparse matrix methods used previously. This makes it practical to refine the detector geometry frequently and thereby maintain accurate calibration without specialized alignment campaigns. Tilts of detector panels out of the plane can be reliably refined, as can the overall distance of the detector in the beam direction. With a simulated test case, the new method produced panel shifts within 7% of the correct values with only one iteration, and produced almost exactly correct shifts after a second iteration. A simulated out-of-plane panel rotation was correctly determined to within 0.001°. Applied to experimental data from an X-ray free-electron laser, the method increased the indexable fraction of frames from 30% to 91% in a single iteration, and to 96% after two further iterations. Computing the geometry updates on the basis of 2060 crystals took only 0.819 s on desktop computing hardware, including the time taken to read the required data from disk. The scaling was found to be very close to linear for up to 100 980 sets of crystal parameters, which took only 78.2 s to process under the same conditions. The method has been applied as part of a real-time feedback system at a synchrotron radiation beamline, in which an out-of-plane detector tilt of 0.04° was detected and corrected. Possible further applications are also described here.

Millepede-II↗

Crystallization and preliminary X-ray analysis of a phosphopentomutase from Bacillus cereus

Phosphopentomutases (PPMs) interconvert D-ribose 5-phosphate and {alpha}-D-ribose 1-phosphate to link glucose and nucleotide metabolism. PPM from Bacillus cereus was overexpressed in Escherichia coli, purified to homogeneity and crystallized. Bacterial PPMs are predicted to contain a di-metal reaction center, but the catalytically relevant metal has not previously been identified. Sparse-matrix crystallization screening was performed in the presence or absence of 50 mM MnCl{sub 2}. This strategy resulted in the formation of two crystal forms from two chemically distinct conditions. The crystals that formed with 50 mM MnCl{sub 2} were more easily manipulated and diffracted to higher resolution. These results suggest that even if the catalytically relevant metal is not known, the crystallization of putative metalloproteins may still benefit from supplementation of the crystallization screens with potential catalytic metals.

59 BASIC BIOLOGICAL SCIENCES↗

Purification, crystallization and preliminary X-ray diffraction analysis of the carbohydrate-binding region of the Streptococcus gordonii adhesin GspB

The carbohydrate-binding region of the bacterial adhesin GspB from Streptococcus gordonii strain M99 (GspB{sub BR}) was expressed in Escherichia coli and purified using affinity and size-exclusion chromatography. Separate sparse-matrix screening of GspB{sub BR} buffered in either 20 mM Tris pH 7.4 or 20 mM HEPES pH 7.5 resulted in different crystallographic behavior such that different precipitants, salts and additives supported crystallization of GspB{sub BR} in each buffer. While both sets of conditions supported crystal growth in space group P2{sub 1}2{sub 1}2{sub 1}, the crystals had distinct unit-cell parameters of a = 33.3, b = 86.7, c = 117.9 {angstrom} for crystal form 1 and a = 34.6, b = 98.3, c = 99.0 {angstrom} for crystal form 2. Additive screening improved the crystals grown in both conditions such that diffraction extended to beyond 2 {angstrom} resolution. A complete data set has been collected to 1.3 {angstrom} resolution with an overall R{sub merge} value of 0.04 and an R{sub merge} value of 0.33 in the highest resolution shell.

59 BASIC BIOLOGICAL SCIENCES↗

Architecture Independent Integrated Early Performance and Energy Estimation

The end of Dennard’s scaling, coupled to the necessity to keep increasing computational performance in constrained power envelopes, is leading developers towards the need to opti- mize applications not only for performance, but also for energy consumption. In such a scenario, solutions that enable estimating performance and energy consumption of a program as early as possible, even during development and without the necessity to access the final target hardware, can become fundamental tools to reduce the design space, the optimization effort, and provide potential opportunities to automate the optimization process itself. In this position paper, we discuss the design of a performance and energy estimation framework based on a retargetable com- piler. Our proposed approach targets the Intermediate Repre- sentation (IR) of the LLVM compiler. The IR based approach enables, after a model is built, estimation from a host different from the target architecture, even considering dynamic infor- mation. We present the rationale behind such a framework, identify opportunities and components readily available (such as the provided static analysis, quick ways to add instrumentation for dynamic profiling, and the multiple backends), and aspects that still need further research efforts (such as more effective non-linear, or machine learning based models). We also discuss a hand-developed case study, based on a simple sparse matrix vector multiplication with a linear model, to motivate the needs for such a framework.

Estimation, Compilers, Performance Analysis↗

Parallelizing the QUDA Library for Multi-GPU Calculations in Lattice Quantum Chromodynamics

Graphics Processing Units (GPUs) are having a transformational effect on numerical lattice quantum chromodynamics (LQCD) calculations of importance in nuclear and particle physics. The QUDA library provides a package of mixed precision sparse matrix linear solvers for LQCD applications, supporting single GPUs based on NVIDIA's Compute Unified Device Architecture (CUDA). This library, interfaced to the QDP++/Chroma framework for LQCD calculations, is currently in production use on the "9g" cluster at the Jefferson Laboratory, enabling unprecedented price/performance for a range of problems in LQCD. Nevertheless, memory constraints on current GPU devices limit the problem sizes that can be tackled. In this contribution we describe the parallelization of the QUDA library onto multiple GPUs using MPI, including strategies for the overlapping of communication and computation. We report on both weak and strong scaling for up to 32 GPUs interconnected by InfiniBand, on which we sustain in excess of 4 Tflops.

43 PARTICLE ACCELERATORS↗

TU-C-17A-05: Dose Domain Optimization of MLC Leaf Patterns for Highly Complicated 4Ï€ IMRT Plans

Purpose: Highly conformal non-coplanar 4π radiotherapy plans typically require more than 20 intensity-modulated fields to deliver. A novel method to calculate multileaf collimator (MLC) leaf patterns is introduced to maximize delivery efficiency, accuracy and plan quality. Methods: 4 GBM patients, with a prescription dose of 59.4 Gy or 60 Gy, were evaluated using the 4π algorithm using 20 beams. The MLC calculation utilized a least square minimization of the dose distribution, with an anisotropic total variation regularization term to encourage piecewise continuity in the fluence maps. Transforming the fluence to the dose domain required multiplying the fluence with a sparse matrix. Exploiting this property made it feasible to solve the problem using CVX, a MATLAB-based convex modeling framework. The fluence was stratified into even step sizes, and the MLC segments, limited to 300, were calculated. The patients studied were replanned using Eclipse with the same beam angles. Results: Compared to the original 4π plan, the stratified 4π plan increased the maximum/mean dose for, in Gy, by 1.0/0.0 (brainstem), 0.5/0.2 (chiasm), 0.0/0.0 (spinal cord), 1.9/0.3 (L eye), 0.7/0.2 (R eye), 0.4/0.4 (L lens), 0.3/0.3 (R lens), 1.0/0.8 (L Optical Nerve), 0.5/0.3 (R Optical Nerve), 0.3/0.2 (L Cochlea), 0.1/0.1 (R Cochlea), 4.6/0.2 (brain), 2.4/0.1 (brain-PTV), 5.1/0.9 (PTV). Compared to Eclipse, which generated an average of 607 segments, the stratified plan reduced (−) or increased (+) the maximum/mean dose, in Gy, by −10.2/−4.1 (brainstem), −10.5/−8.9 (chiasm), +0.0/−0.1 (spinal cord), −4.9/−3.4 (L eye), −4.1/−2.5 (R eye), −2.8/−2.7 (L lens), −2.1/−1.9 (R lens), −7.6/−6.5 (L Optical Nerve), −8.9/−6.1 (R Optical Nerve), −1.3/−1.9 (L Cochlea), −1.8/−1.8 (R Cochlea), +1.7/−2.1 (brain), +3.2/−2.6 (brain-PTV), +1.8/+0.3 Gy (PTV. The stratified plan was also more homogeneous in the PTV. Conclusion: This novel solver can transform complicated fluence maps into significantly fewer deliverable MLC segments than the commercial system while achieving superior dosimetry. Funding support partially contributed by Varian.

60 APPLIED LIFE SCIENCES↗

Parallel implementation and performance optimization of the configuration-interaction method

The configuration-interaction (CI) method, long a popular approach to describe quantum many-body systems, is cast as a very large sparse matrix eigenpair problem with matrices whose dimension can exceed one billion. Such formulations place high demands on memory capacity and memory bandwidth - - two quantities at a premium today. In this paper, we describe an efficient, scalable implementation, BIGSTICK, which, by factorizing both the basis and the interaction into two levels, can reconstruct the nonzero matrix elements on the fly, reduce the memory requirements by one or two orders of magnitude, and enable researchers to trade reduced resources for increased computational time. We optimize BIGSTICK on two leading HPC platforms - - the Cray XC30 and the IBM Blue Gene/Q. Specifically, we not only develop an empirically-driven load balancing strategy that can evenly distribute the matrix-vector multiplication across 256K threads, we also developed techniques that improve the performance of the Lanczos reorthogonalization. Combined, these optimizations improved performance by 1.3-8× depending on platform and configuration.

97 MATHEMATICS AND COMPUTING↗