Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “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 505 records · Page 28

Superoxide produced in the matrix of mitochondria enhances methylmercury toxicity in human neuroblastoma cells

The mechanism of intracellular metabolism of methylmercury (MeHg) is not fully known. It has been shown that superoxide (O{sub 2}·{sup −}), the proximal reactive oxygen species (ROS) generated by mitochondria, is responsible for MeHg demethylation. Here, we investigated the impact of different mitochondrial respiratory inhibitors, namely rotenone and antimycin A, on the O{sub 2}·{sup −} mediated degradation of MeHg in human neuroblastoma cells SH-K-SN. We also utilized paraquat (PQ) which generates O{sub 2}·{sup −} in the mitochondrial matrix. We found that the cleavage of the carbon-metal bond in MeHg was highly dependent on the topology of O{sub 2}·{sup −} production by mitochondria. Both rotenone and PQ, which increase O{sub 2}·{sup −} in the mitochondrial matrix at a dose-dependent manner, enhanced the conversion of MeHg to inorganic mercury (iHg). Surprisingly, antimycin A, which prompts emission of O{sub 2}·{sup −} into the intermembrane space, did not have the same effect even though antimycin A induced a dose dependent increase in O{sub 2}·{sup −} emission. Rotenone and PQ also enhanced the toxicity of sub-toxic doses (0.1 μM) MeHg which correlated with the accumulation of iHg in mitochondria and depletion of mitochondrial protein thiols. Taken together, our results demonstrate that MeHg degradation is mediated by mitochondrial O{sub 2}·{sup −}, specifically within the matrix of mitochondria when O{sub 2}·{sup −} is in adequate supply. Our results also show that O{sub 2}·{sup −} amplifies MeHg toxicity specifically through its conversion to iHg and subsequent interaction with protein cysteine thiols (R-SH). The implications of our findings in mercury neurotoxicity are discussed herein. - Highlights: • Superoxide produced in the matrix of mitochondria degrades MeHg. • Superoxide produced in intermembrane space does not degrade MeHg. • Matrix-generated superoxide enhances Hg toxicity by converting MeHg to iHg.

60 APPLIED LIFE SCIENCES↗

Estrogen depletion alters osteogenic differentiation and matrix production by osteoblasts in vitro

Highlights: • Estrogen depletion accelerates osteoblast differentiation in static culture. • Mechanical stimulation exacerbated the effects of estrogen depletion on osteoblast differentiation. • Estrogen depletion alters osteoblasts capacity to produce a good quality matrix. Recent studies have revealed that the effects of estrogen deficiency are not restricted to osteoclasts and bone resorption, but that bone matrix composition is altered and osteoblasts exhibit an impaired response to mechanical stimulation. In this study, we test the hypothesis that estrogen depletion alters osteogenic differentiation and matrix production by mechanically stimulated osteoblasts in vitro. MC3T3-E1 cells were pre-treated with estrogen for 14 days, after which estrogen was withdrawn or inhibited with Fulvestrant up to 14 days. Fluid shear stress (FSS) was applied using an orbital shaker. Under estrogen depletion in static culture, osteogenic marker (ALP) and gene expression (Runx2) were decreased at 2 and after 7 days of estrogen depletion, respectively. In addition, up to 7 day the inhibition of the estrogen receptor significantly decreased fibronectin expression (FN1) under static conditions. Under estrogen depletion and daily mechanical stimulation, changes in expression of Runx2 occurred earlier (4 days) and by 14 days, changes in matrix production (Col1a1) were reported. We propose that changes in osteoblast differentiation and impaired matrix production during estrogen depletion may contribute to the altered quality of the bone and act as a contributing factor to increased bone fragility in postmenopausal osteoporosis.

60 APPLIED LIFE SCIENCES↗

Is the Matrix Completion of Reduced Density Matrices Unique?

Reduced density matrices are central to describing observables in many-body quantum systems. In electronic structure theory, the two-particle reduced density matrix (2-RDM) suffices to determine the energy and other key properties. Recent work has used matrix completion, leveraging the low-rank structure of RDMs and approximate theoretical models, to reconstruct the 2-RDM from partial data and thus reduce the computational cost. However, matrix completion is, in general, an under-determined problem. Revisiting Rosina’s theorem (Rosina, M. Queen’s Papers on Pure and Applied Mathematics , 1968, No. 11, 369), we here show that the matrix completion is unique under certain conditions, identifying the subset of 2-RDM elements that enables its exact reconstruction from incomplete information. Building on this, we introduce a hybrid quantum–stochastic algorithm that achieves exact matrix completion, demonstrated through applications to the Fermi–Hubbard model.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Matrix-Assisted Ionization of Molecular Uranium Species

Matrix-assisted ionization (MAI) demonstrates high sensitivity for a variety of organic compounds; however, few studies have reported the application of MAI for the detection and characterization of inorganic analytes. Trace-level uranium analysis is important in the realms of nuclear forensics, nuclear safeguards, and environmental monitoring. Traditional mass spectrometry methods employed in these fields require combinations of extensive laboratory chemistry sample preparation and destructive ionization methods. There has been recent interest in exploring ambient mass spectrometry methods that enable timely sample analysis and higher sensitivity than what is attainable by field-portable radiation detectors. Rapid characterization of uranium at nanogram levels is demonstrated in this study using MAI techniques. Mass spectra were collected on an atmospheric pressure mass spectrometer for solutions of uranyl nitrate, uranyl chloride, uranyl acetate, and uranyl oxalate utilizing 3-nibrobenzonitrile as the ionization matrix. The uranyl complexes investigated were detectable, and the chemical speciation was preserved. Sample analysis was accomplished in a matter of seconds, and limits of detection of 5 ng of uranyl nitrate, 10 ng of uranyl oxalate, 100 ng of uranyl chloride, and 200 ng of uranyl acetate were achieved. The observed gas-phase speciation was similar to negative-ion electrospray ionization of uranyl compounds with notable differences. Six matrix-derived ions were detected in all negative-ion mass spectra, and some of these ions formed adducts with the uranyl analyte. Subsequent analysis of the matrix suggests that these molecules are not matrix contaminants and are instead created during the ionization process.

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗

A de novo matrix for macroscopic living materials from bacteria

Engineered living materials (ELMs) embed living cells in a biopolymer matrix to create materials with tailored functions. While bottom-up assembly of macroscopic ELMs with a de novo matrix would offer the greatest control over material properties, we lack the ability to genetically encode a protein matrix that leads to collective self-organization. Here we report growth of ELMs from Caulobacter crescentus cells that display and secrete a self-interacting protein. This protein formed a de novo matrix and assembled cells into centimeter-scale ELMs. Discovery of design and assembly principles allowed us to tune the composition, mechanical properties, and catalytic function of these ELMs. This work provides genetic tools, design and assembly rules, and a platform for growing ELMs with control over both matrix and cellular structure and function.

59 BASIC BIOLOGICAL SCIENCES↗

Toward Active-Matrix Lab-On-A-Chip: Programmable Electrofluidic control Enaled by Arrayed Oxide Thin Film Transistors

Agile micro- and nano-fluidic control is critical to numerous life science and chemical science synthesis as well as kinetic and thermodynamic studies. To this end, we have demonstrated the use of thin film transistor arrays as an active matrix addressing method to control an electrofluidic array. Because the active matrix method minimizes the number of control lines necessary (m + n lines for the m x n element array), the active matrix addressing method integrated with an electrofluidic platform can be a significant breakthrough for complex electrofluidic arrays (increased size or resolution) with enhanced function, agility and programmability. An amorphous indium gallium zinc oxide (a-IGZO) semiconductor active layer is used because of its high mobility of 1-15 cm{sup 2} V{sup -1} s{sup -1}, low-temperature processing and transparency for potential spectroscopy and imaging. Several electrofluidic functionalities are demonstrated using a simple 2 x 5 electrode array connected to a 2 x 5 IGZO thin film transistor array with the semiconductor channel width of 50 {mu}m and mobility of 6.3 cm{sup 2} V{sup -1} s{sup -1}. Additionally, using the TFT device characteristics, active matrix addressing schemes are discussed as the geometry of the electrode array can be tailored to act as a storage capacitor element. Finally, requisite material and device parameters are discussed in context with a VGA scale active matrix addressed electrofluidic platform.

59 BASIC BIOLOGICAL SCIENCES↗

A solvent replenishment solution for managing evaporation of biochemical reactions in air-matrix digital microfluidics devices

Digital microfluidics (DMF) is a powerful technique for sample preparation and analysis for a broad range of biological and chemical applications. In many cases, it is desirable to carry out DMF on an open surface, such that the matrix surrounding the droplets is ambient air. However, the utility of the air-matrix DMF format has been severely limited by problems with droplet evaporation, especially when the droplet-based biochemical reactions require high temperatures for long periods of time. We present a simple solution for managing evaporation in air-matrix DMF: just-in-time replenishment of the reaction volume using droplets of solvent. We demonstrate that this solution enables DMF-mediated execution of several different biochemical reactions (RNA fragmentation, first-strand cDNA synthesis, and PCR) over a range of temperatures (4–95 °C) and incubation times (up to 1 h or more) without use of oil, humidifying chambers, or off-chip heating modules. Reaction volumes and temperatures were maintained roughly constant over the course of each experiment, such that the reaction kinetics and products generated by the air-matrix DMF device were comparable to those of conventional benchscale reactions. As a result, this simple yet effective solution for evaporation management is an important advance in developing air-matrix DMF for a wide variety of new, high-impact applications, particularly in the biomedical sciences.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

The asymptotic behaviour of the scattering matrix in a neighbourhood of the endpoints of a spectral gap

The behaviour of the scattering matrix is investigated as the spectral parameter approaches an endpoint of a spectral gap of a quantum waveguide from the inside or the outside. The waveguide has two sleeves, one is cylindrical and the other periodic. When the spectral parameter traverses the spectral gap, the scattering matrix is reshaped because the number of waves inside and outside the gap is different. Notwithstanding, the smaller scattering matrix (in size) is transformed continuously into an identical block in the bigger scattering matrix and, in addition, the latter takes block diagonal form in the limit at the endpoint of the gap, that is, at the spectral threshold. The unexpected phenomena are related to the other block. It is shown that in the limit this block can only take certain values at the threshold, and taking one or other of these values depends on the structure of the continuous spectrum and also on the structure of the subspace of 'almost standing' waves at the threshold, which are solutions of the homogeneous problem that transfer no energy to infinity. A criterion for the existence of such solutions links the dimension of this subspace to the multiplicity of the eigenvalue −1 of the threshold scattering matrix. Asymptotic formulae are obtained, which show, in particular, that the phenomenon of anomalous scattering of high-amplitude waves at near-threshold frequencies, discovered by Weinstein in a special acoustic problem, also occurs in periodic waveguides. Bibliography: 38 titles. (paper)

97 MATHEMATICS AND COMPUTING↗

Hot Hydrogen Testing of W-Coated UN Kernels in a Mo30W Matrix

Ceramic uranium mononitride (UN) is being considered as a reactor fuel for nuclear thermal propulsion. To avoid or reduce the dissociation of UN at the high temperatures needed, embedding it in a metallic matrix (cermet) has been proposed. To assess the viability of this concept, hot hydrogen testing of tungsten-coated UN kernels embedded in a Mo-30 wt% W (Mo30W) alloy matrix has been performed at temperatures from 1800°C to 2300°C. Both the isolated kernels and kernels consolidated by spark plasma sintering in the Mo30W matrix were tested. In addition to direct observations and mass loss measurements, the samples were analyzed by X-ray diffraction (XRD) and scanning electron microscopy (SEM)/energy dispersive X-ray spectroscopy (EDS) after each run. The decomposition of UN started at 1800°C despite the coating and matrix, and increased at 2000°C. Uranium seeped through the tungsten grain boundaries of the coating at all temperatures. The consolidated sample expanded irregularly at 2000°C through the formation of voids, and SEM/EDS analysis showed uranium-containing veins in the matrix consisting of U 2 Mo according to the XRD data. The observed pore generation at 2000°C was explained by the formation of water vapor from residual oxides and diffused hydrogen. At 2200°C and above, both the kernels and the consolidated samples melted through the formation of uranium or low–melting point uranium-molybdenum alloys.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗

Quantitative comparison of opacities calculated using the R -matrix and distorted-wave methods: Fe xvii

ABSTRACT We present here a detailed calculation of opacities for Fe xvii at the physical conditions corresponding to the base of the Solar convection zone. Many ingredients are involved in the calculation of opacities. We review the impact of each ingredient on the final monochromatic and mean opacities (Rosseland and Planck). The necessary atomic data were calculated with the R-matrix and the distorted-wave (DW) methods. We study the effect of broadening, of resolution, of the extent of configuration sets and of configuration interaction to understand the differences between several theoretical predictions as well as the existing large disagreement with measurements. New Dirac R-matrix calculations including all configurations up to the n = 4, 5, and 6 complexes have been performed as well as corresponding Breit–Pauli DW calculations. The DW calculations have been extended to include autoionizing initial levels. A quantitative contrast is made between comparable DW and R-matrix models. We have reached self-convergence with n = 6 R-matrix and DW calculations. Populations in autoionizing initial levels contribute significantly to the opacities and should not be neglected. The R-matrix and DW results are consistent under the similar treatment of resonance broadening. The comparison with the experiment shows a persistent difference in the continuum while the filling of the windows shows some improvement. This study defines our path to the next generation of opacities and opacity tables for stellar modelling.

Delahaye, F.↗

The DYRKP1 kinase regulates cell wall degradation in Chlamydomonas by inducing matrix metalloproteinase expression

Abstract The cell wall of plants and algae is an important cell structure that protects cells from changes in the external physical and chemical environment. This extracellular matrix, composed of polysaccharides and glycoproteins, must be constantly remodeled throughout the life cycle. However, compared to matrix polysaccharides, little is known about the mechanisms regulating the formation and degradation of matrix glycoproteins. We report here that a plant kinase belonging to the dual-specificity tyrosine phosphorylation-regulated kinase (DYRKP1) family present in all eukaryotes regulates cell wall degradation after mitosis of Chlamydomonas reinhardtii by inducing the expression of matrix metalloproteinases. Without DYRKP1, daughter cells cannot disassemble parental cell walls and remain trapped inside for more than 10 days. On the other hand, the dual-specificity tyrosine phosphorylation-regulated kinase complementation lines show normal degradation of the parental cell wall. Transcriptomic and proteomic analyses indicate a marked downregulation of MMP gene expression and accumulation, respectively, in the dyrkp1 mutants. The mutants deficient in matrix metalloproteinases retain palmelloid structures for a longer time than the background strain, like dyrkp1 mutants. Our findings show that dual-specificity tyrosine phosphorylation-regulated kinase, by ensuring timely MMP expression, enables the successful execution of the cell cycle. Altogether, this study provides insight into the life cycle regulation in plants and algae.

Kim, Minjae (ORCID:0000000223561295)↗

Geometric quench in the fractional quantum Hall effect: Exact solution in quantum Hall matrix models and comparison with bimetric theory

In this work, we investigate the recently introduced geometric quench protocol for fractional quantum Hall (FQH) states within the framework of exactly solvable quantum Hall matrix models. In the geometric quench protocol, a FQH state is subjected to a sudden change in the ambient geometry, which introduces anisotropy into the system. We formulate this quench in the matrix models and then we solve exactly for the postquench dynamics of the system and the quantum fidelity (Loschmidt echo) of the postquench state. Next, we explain how to define a spin-2 collective variable g ' a b ( t ) in the matrix models, and we show that for a weak quench (small anisotropy), the dynamics of g ' a b ( t ) agrees with the dynamics of the intrinsic metric governed by the recently discussed bimetric theory of FQH states. We also find a modification of the bimetric theory such that the predictions of the modified bimetric theory agree with those of the matrix model for arbitrarily strong quenches. Finally, we introduce a class of higher-spin collective variables for the matrix model, which are related to generators of the W ∞ algebra, and we show that the geometric quench induces nontrivial dynamics for these variables.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Nonadiabatic molecular dynamics with decoherence and detailed balance under a density matrix ensemble formalism

The mixed quantum-classical nonadiabatic molecular dynamics (NAMD) is a powerful tool to study many phenomena, especially ultrafast carrier transport and cooling. Carrier decoherence and detailed balance are two major issues in NAMD. So far, there is no computationally inexpensive approach to incorporate both effects. While the decoherence effect can be easily included in the state density matrix formalism and the detailed balance can be included in surface hopping or the wave function collapse approach, it is difficult to include both of them in a unified formalism. In this work we introduce a state density matrix formalism (referred to as P-matrix) including both the decoherence and detailed balance effects for NAMD. This method is able to explicitly treat the decoherence between different pairs of adiabatic states. Moreover, the off-diagonal density matrix elements are divided into two parts, corresponding to energy-increasing and energy-decreasing transitions. The detailed balance is then enforced by a Boltzmann factor applied to the energy-increasing transition part. The P-matrix formalism is applied to study hot-hole cooling and transfer processes in Si quantum dot (QD) systems. The calculated hot-carrier relaxation time is consistent with experiments. In a QD-pair system, the hot-hole cooling time shows weak dependence on the QD spacing. However, the hot-carrier transfer rate from one QD to another is found to decrease exponentially with the QD-QD distance. When the QD spacing is small ( ~ 1 nm), the hot-carrier transfer can be very efficient. It is also shown that the explicit treatment of decoherence time is important in order to treat this hot-carrier transfer correctly.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Shadow poles in the alternative parametrization of R-matrix theory

In this work, we discover new, hitherto unknown, shadow poles in Brune’s alternative parametrization of R-matrix theory [C. R. Brune, Phys. Rev. C 66, 044611 (2002)]. Where these poles are, and how many, depends on how one continues R-matrix operators to complex wavenumbers (specially the shift S and penetration P functions). This has little consequence for the exact R-matrix formalism (past the last energy threshold), as we show one can still always fully reconstruct the scattering matrix with only the previously known alternative parameters (poles and corresponding resonance widths), for which there were as many poles as the number of levels Nλ. However, we generalize the alternative parametrization to the Reich-Moore formalism, and show that the choice of continuation is now critical as it changes the alternative parameters values (poles and residue widths are now complex). In order to establish nuclear libraries with alternative parameters, the nuclear community will thus have to decide what convention to adopt. We argue in favor of analytical continuation (against the legacy Lane and Thomas approach) in a follow-up article [P. Ducru, Phys. Rev. C, submitted (2020)]. We observe the first evidence of shadow poles in the alternative parametrization of R-matrix theory in isotope xenon 134 Xe spin-parity group J π = 1=2 (–) , and show how they indeed depend on the choice of continuation to complex wavenumbers.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Windowed multipole representation of R -matrix cross sections

Nuclear cross sections are basic inputs to any nuclear computation. Campaigns of experiments are fitted with the parametric R-matrix model of quantum nuclear interactions, and the resulting cross sections are documented—both pointwise and as resonance parameters (with uncertainties)—in standard evaluated nuclear data libraries (ENDF, JEFF, BROND, JENDL, CENDL, TENDL): these constitute our common knowledge of fundamental low-energy nuclear cross sections. In the past decade, a collaborative effort has been deployed to establish a new nuclear cross-section library format—the Windowed Multipole Library—with the goal of considerably reducing the computational cost of cross-section calculations in nuclear transport simulations. This work lays the theoretical foundations underpinning these efforts. From general R-matrix scattering theory, we derive the windowed multipole representation of nuclear cross sections. Though physically and mathematically equivalent to R-matrix cross sections, the windowed multipole representation is particularly well suited for subsequent temperature treatment of angle-integrated cross sections, in particular Doppler broadening, which is the averaging of cross sections over the thermal motion of the target atoms. Doppler broadening is of critical importance in neutron transport applications, as it ensures the stability of many nuclear reactors (negative thermal reactivity). Yet, Doppler broadening of nuclear cross sections has been a considerable bottleneck for nuclear transport computations, often requiring memory-costly pretabulations. We show that the windowed multipole representation can perform accurate Doppler broadening analytically (up to the first reaction threshold), from which we derive cross-section temperature derivatives to any order—all computable on the fly (without precalculations stored in memory). Furthermore, we here establish a way of converting the R-matrix resonance parameters uncertainty (covariance matrices) into windowed multipole parameters uncertainty. We show that generating stochastic nuclear cross sections by sampling from the resulting windowed multipole covariance matrix can reproduce the cross-section uncertainty in the original nuclear data file. The windowed multipole representation is therefore a novel nuclear physics formalism able to generate Doppler broadened stochastic nuclear cross sections on the fly, unlocking breakthrough computational gains for nuclear computations. Through this foundational paper, we hope to make the windowed multipole representation accessible, reproducible, and usable for the nuclear physics community, as well as provide the theoretical basis for future research on expanding its capabilities.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Determining the heavy seesaw neutrino mass matrix from low-energy parameters

We explore how the seesaw sector in neutrino mass models may be constrained through symmetries to be completely determined in terms of low-energy mass, mixing angle and CP-violating phase observables. The key ingredients are intrafamily symmetries to determine the neutrino Dirac mass matrix in terms of the charged-lepton or quark mass matrices, together with interfamily or flavor symmetries to determine diagonalization matrices. Implications for leptogenesis and collider detection of heavy neutral leptons are discussed. We show that leptogenesis can succeed in small regions of parameter space for the case where the neutrino Dirac mass matrix equals the up-quark mass matrix. The model where the neutrino Dirac mass matrix equals the charged-lepton mass matrix can yield a heavy neutral lepton as light as about 1 TeV, but detecting such a particle will be difficult.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Chiral matrix model of the semi-QGP in QCD

Previously, a matrix model of the region near the transition temperature, in the “semi”quark gluon plasma, was developed for the theory of SU(3) gluons without quarks. In this paper we develop a chiral matrix model applicable to QCD by including dynamical quarks with 2+1 flavors. This requires adding a nonet of scalar fields, with both parities, and coupling these to quarks through a Yukawa coupling, y. Treating the scalar fields in mean field approximation, the effective Lagrangian is computed by integrating out quarks to one loop order. As is standard, the potential for the scalar fields is chosen to be symmetric under the flavor symmetry of SU (3) L × SU(3) R × Z (3) A , except for a term linear in the current quark mass, m qk . In addition, at a nonzero temperature T it is necessary to add a new term, ~ m qk T 2 . The parameters of the gluon part of the matrix model are identical to those for the pure glue theory without quarks. The parameters in the chiral matrix model are fixed by the values, at zero temperature, of the pion decay constant and the masses of the pions, kaons, η , and η' . The temperature for the chiral crossover at T$χ$ = 155 MeV is determined by adjusting the Yukawa coupling y . We find reasonable agreement with the results of numerical simulations on the lattice for the pressure and related quantities. In the chiral limit, besides the divergence in the chiral susceptibility there is also a milder divergence in the susceptibility between the Polyakov loop and the chiral order parameter, with critical exponent β $-$ 1 . We compute derivatives with respect to a quark chemical potential to determine the susceptibilities for baryon number, the $χ$ 2n . Especially sensitive tests are provided by $χ$ 4 $-$ $χ$ 2 and by $χ$ 6 , which changes in sign about T$χ$ . In conclusion, the behavior of the susceptibilities in the chiral matrix model strongly suggests that as the temperature increases from T$χ$ , that the transition to deconfinement is significantly quicker than indicated by the measurements of the (renormalized) Polyakov loop on the lattice.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

A Work-Efficient Parallel Sparse Matrix-Sparse Vector Multiplication Algorithm

We design and develop a work-efficient multithreaded algorithm for sparse matrix-sparse vector multiplication (SpMSpV) where the matrix, the input vector, and the output vector are all sparse. SpMSpV is an important primitive in the emerging GraphBLAS standard and is the workhorse of many graph algorithms including breadth-first search, bipartite graph matching, and maximal independent set. As thread counts increase, existing multithreaded SpMSpV algorithms can spend more time accessing the sparse matrix data structure than doing arithmetic. Our shared-memory parallel SpMSpV algorithm is work efficient in the sense that its total work is proportional to the number of arithmetic operations required. The key insight is to avoid each thread individually scan the list of matrix columns. Our algorithm is simple to implement and operates on existing column-based sparse matrix formats. It performs well on diverse matrices and vectors with heterogeneous sparsity patterns. A high-performance implementation of the algorithm attains up to 15x speedup on a 24-core Intel Ivy Bridge processor and up to 49x speedup on a 64-core Intel KNL manycore processor. In contrast to implementations of existing algorithms, the performance of our algorithm is sustained on a variety of different input types include matrices representing scale-free and high-diameter graphs.

97 MATHEMATICS AND COMPUTING↗