Publications

Abstract not provided.

Abstract not provided.

A graph is a mathematical representation of a network; we say it consists of a set of vertices, which are connected by edges. Graphs have numerous applications in various fields, as they can model all sorts of connections, processes, or relations. For example, graphs can model intricate transit systems or the human nervous system. However, graphs that are large or complicated become difficult to analyze. This is why there is an increased interest in the area of graph partitioning, reducing the size of the graph into multiple partitions. For example, partitions of a graph representing a social network might help identify clusters of friends or colleagues. Graph partitioning is also a widely used approach to load balancing in parallel computing. The partitioning of a graph is extremely useful to decompose the graph into smaller parts and allow for easier analysis. There are different ways to solve graph partitioning problems. For this work, we focus on a spectral partitioning method which forms a partition based upon the eigenvectors of the graph Laplacian (details presented in Acer, et. al.). This method uses the LOBPCG algorithm to compute these eigenvectors. LOBPCG can be accelerated by an operator called a preconditioner. For this internship, we evaluate a randomized Cholesky (rchol) preconditioner for its effectiveness on graph partitioning problems with LOBPCG. We compare it with two standard preconditioners: Jacobi and Incomplete Cholesky (ichol). This research was conducted from August to December 2021 in conjunction with Sandia National Laboratories.

We propose a novel statistical inference paradigm for zero-inflated multiway count data that dispenses with the need to distinguish between true and false zero counts. Our approach ignores all zero entries and applies zero-truncated Poisson regression on the positive counts. Inference is accomplished via tensor completion that imposes low-rank structure on the Poisson parameter space. Our main result shows that an $\textit{N}$-way rank-R parametric tensor 𝓜 ϵ (0, ∞)^{$I$Χ∙∙∙Χ$I$} generating Poisson observations can be accurately estimated from approximately $IR^2 \text{log}^2_2(I)$ non-zero counts for a nonnegative canonical polyadic decomposition. Several numerical experiments are presented demonstrating that our zero-truncated paradigm is comparable to the ideal scenario where the locations of false zero counts are known $\textit{a priori}$.

Nuclear spins were among the first physical platforms to be considered for quantum information processing1,2, because of their exceptional quantum coherence3 and atomic-scale footprint. However, their full potential for quantum computing has not yet been realized, owing to the lack of methods with which to link nuclear qubits within a scalable device combined with multi-qubit operations with sufficient fidelity to sustain fault-tolerant quantum computation. Here we demonstrate universal quantum logic operations using a pair of ion-implanted 31P donor nuclei in a silicon nanoelectronic device. A nuclear two-qubit controlled-Z gate is obtained by imparting a geometric phase to a shared electron spin4, and used to prepare entangled Bell states with fidelities up to 94.2(2.7)%. The quantum operations are precisely characterized using gate set tomography (GST)5, yielding one-qubit average gate fidelities up to 99.95(2)%, two-qubit average gate fidelity of 99.37(11)% and two-qubit preparation/measurement fidelities of 98.95(4)%. These three metrics indicate that nuclear spins in silicon are approaching the performance demanded in fault-tolerant quantum processors6. We then demonstrate entanglement between the two nuclei and the shared electron by producing a Greenberger–Horne–Zeilinger three-qubit state with 92.5(1.0)% fidelity. Because electron spin qubits in semiconductors can be further coupled to other electrons7–9 or physically shuttled across different locations10,11, these results establish a viable route for scalable quantum information processing using donor nuclear and electron spins.

Anomalous behavior is ubiquitous in subsurface solute transport due to the presence of high degrees of heterogeneity at different scales in the media. Although fractional models have been extensively used to describe the anomalous transport in various subsurface applications, their application is hindered by computational challenges. Simpler nonlocal models characterized by integrable kernels and finite interaction length represent a computationally feasible alternative to fractional models; yet, the informed choice of their kernel functions still remains an open problem. We propose a general data-driven framework for the discovery of optimal kernels on the basis of very small and sparse data sets in the context of anomalous subsurface transport. Using spatially sparse breakthrough curves recovered from fine-scale particle-density simulations, we learn the best coarse-scale nonlocal model using a nonlocal operator regression technique. Predictions of the breakthrough curves obtained using the optimal nonlocal model show good agreement with fine-scale simulation results even at locations and time intervals different from the ones used to train the kernel, confirming the excellent generalization properties of the proposed algorithm. A comparison with trained classical models and with black-box deep neural networks confirms the superiority of the predictive capability of the proposed model.

Krylov subspace recycling is a powerful tool when solving a long series of large, sparse linear systems that change only slowly over time. In PDE constrained shape optimization, these series appear naturally, as typically hundreds or thousands of optimization steps are needed with only small changes in the geometry. In this setting, however, applying Krylov subspace recycling can be a difficult task. As the geometry evolves, in general, so does the finite element mesh defined on or representing this geometry, including the numbers of nodes and elements and element connectivity. This is especially the case if re-meshing techniques are used. As a result, the number of algebraic degrees of freedom in the system changes, and in general the linear system matrices resulting from the finite element discretization change size from one optimization step to the next. Changes in the mesh connectivity also lead to structural changes in the matrices. In the case of re-meshing, even if the geometry changes only a little, the corresponding mesh might differ substantially from the previous one. Obviously, this prevents any straightforward mapping of the approximate invariant subspace of the linear system matrix (the focus of recycling in this work) from one optimization step to the next; similar problems arise for other selected subspaces. In this paper, we present an algorithm to map an approximate invariant subspace of the linear system matrix for the previous optimization step to an approximate invariant subspace of the linear system matrix for the current optimization step, for general meshes. This is achieved by exploiting the map from coefficient vectors to finite element functions on the mesh, combined with interpolation or approximation of functions on the finite element mesh. We demonstrate the effectiveness of our approach numerically with several proof of concept studies for a specific meshing technique.

Numerical simulations of Greenland and Antarctic ice sheets involve the solution of large-scale highly nonlinear systems of equations on complex shallow geometries. This work is concerned with the construction of Schwarz preconditioners for the solution of the associated tangent problems, which are challenging for solvers mainly because of the strong anisotropy of the meshes and wildly changing boundary conditions that can lead to poorly constrained problems on large portions of the domain. Here, two-level generalized Dryja-Smith-Widlund (GDSW)-type Schwarz preconditioners are applied to different land ice problems, i.e., a velocity problem, a temperature problem, as well as the coupling of the former two problems. We employ the message passing interface (MPI)- parallel implementation of multilevel Schwarz preconditioners provided by the package FROSch (fast and robust Schwarz) from the Trilinos library. The strength of the proposed preconditioner is that it yields out-of-the-box scalable and robust preconditioners for the single physics problems. To the best of our knowledge, this is the first time two-level Schwarz preconditioners have been applied to the ice sheet problem and a scalable preconditioner has been used for the coupled problem. The preconditioner for the coupled problem differs from previous monolithic GDSW preconditioners in the sense that decoupled extension operators are used to compute the values in the interior of the subdomains. Several approaches for improving the performance, such as reuse strategies and shared memory OpenMP parallelization, are explored as well. In our numerical study we target both uniform meshes of varying resolution for the Antarctic ice sheet as well as nonuniform meshes for the Greenland ice sheet. We present several weak and strong scaling studies confirming the robustness of the approach and the parallel scalability of the FROSch implementation. Among the highlights of the numerical results are a weak scaling study for up to 32 K processor cores (8 K MPI ranks and 4 OpenMP threads) and 566 M degrees of freedom for the velocity problem as well as a strong scaling study for up to 4 K processor cores (and MPI ranks) and 68 M degrees of freedom for the coupled problem.

Deep operator learning has emerged as a promising tool for reduced-order modelling and PDE model discovery. Leveraging the expressive power of deep neural networks, especially in high dimensions, such methods learn the mapping between functional state variables. While proposed methods have assumed noise only in the dependent variables, experimental and numerical data for operator learning typically exhibit noise in the independent variables as well, since both variables represent signals that are subject to measurement error. In regression on scalar data, failure to account for noisy independent variables can lead to biased parameter estimates. With noisy independent variables, linear models fitted via ordinary least squares (OLS) will show attenuation bias, wherein the slope will be underestimated. In this work, we derive an analogue of attenuation bias for linear operator regression with white noise in both the independent and dependent variables, showing that the norm upper bound of the operator learned via OLS decreases with increasing noise in the independent variable. In the nonlinear setting, we computationally demonstrate underprediction of the action of the Burgers operator in the presence of noise in the independent variable. We propose error-in-variables (EiV) models for two operator regression methods, MOR-Physics and DeepONet, and demonstrate that these new models reduce bias in the presence of noisy independent variables for a variety of operator learning problems. Considering the Burgers operator in 1D and 2D, we demonstrate that EiV operator learning robustly recovers operators in high-noise regimes that defeat OLS operator learning. We also introduce an EiV model for time-evolving PDE discovery and show that OLS and EiV perform similarly in learning the Kuramoto-Sivashinsky evolution operator from corrupted data, suggesting that the effect of bias in OLS operator learning depends on the regularity of the target operator.

Magnetic skyrmions are nanoscale whirls of magnetism that can be propagated with electrical currents. The repulsion between skyrmions inspires their use for reversible computing based on the elastic billiard ball collisions proposed for conservative logic in 1982. In this letter, we evaluate the logical and physical reversibility of this skyrmion logic paradigm, as well as the limitations that must be addressed before dissipation-free computation can be realized.

Abstract not provided.

Software sustainability is critical for Computational Science and Engineering (CSE) software. Measuring sustainability is challenging because sustainability consists of many attributes. One factor that impacts software sustainability is the complexity of the source code. This paper introduces an approach for utilizing complexity data, with a focus on hotspots of and changes in complexity, to assist developers in performing code reviews and inform project teams about longer-term changes in sustainability and maintainability from the perspective of cyclomatic complexity. We present an analysis of data associated with four real-world pull requests to demonstrate how the metrics may help guide and inform the code review process and how the data can be used to measure changes in complexity over time.

The National Academy of Sciences, Engineering, and Medicine (NASEM) defines reproducibility as 'obtaining consistent computational results using the same input data, computational steps, methods, code, and conditions of analysis,' and replicability as 'obtaining consistent results across studies aimed at answering the same scientific question, each of which has obtained its own data' [1]. Due to an increasing number of applications of artificial intelligence and machine learning (AI/ML) to fields such as healthcare and digital medicine, there is a growing need for verifiable AI/ML results, and therefore reproducible research and replicable experiments. This paper establishes examples of irreproducible AI/ML applications to medical sciences and quantifies the variance of common AI/ML models (Artificial Neural Network, Naive Bayes classifier, and Random Forest classifiers) for tasks on medical data sets.

Measurements that occur within the internal layers of a quantum circuit—midcircuit measurements—are a useful quantum-computing primitive, most notably for quantum error correction. Midcircuit measurements have both classical and quantum outputs, so they can be subject to error modes that do not exist for measurements that terminate quantum circuits. Here we show how to characterize midcircuit measurements, modeled by quantum instruments, using a technique that we call quantum instrument linear gate set tomography (QILGST). We then apply this technique to characterize a dispersive measurement on a superconducting transmon qubit within a multiqubit system. By varying the delay time between the measurement pulse and subsequent gates, we explore the impact of residual cavity photon population on measurement error. QILGST can resolve different error modes and quantify the total error from a measurement; in our experiment, for delay times above 1000ns we measure a total error rate (i.e., half diamond distance) of ϵ⋄=8.1±1.4%, a readout fidelity of 97.0±0.3%, and output quantum-state fidelities of 96.7±0.6% and 93.7±0.7% when measuring 0 and 1, respectively.

This paper applies sensitivity and uncertainty analysis to compare two model alternatives for fuel matrix degradation for performance assessment of a generic crystalline repository. The results show that this model choice has little effect on uncertainty in the peak 129I concentration. The small impact of this choice is likely due to the higher importance of uncertainty in the instantaneous release fraction and differences in epistemic uncertainty between the alternatives.

Multi-model Monte Carlo methods have been illustrated to be an efficient and accurate alternative to standard Monte Carlo (MC) in the model-based propagation of uncertainty in entry, descent, and landing (EDL) applications. These multi-model MC methods fuse predictions from low-fidelity models with the high-fidelity EDL model of interest to produce unbiased statistics with a fraction of the computational cost. The accuracy and efficiency of the multi-model MC methods are dependent upon the magnitude of correlations of the low-fidelity models with the high-fidelity model, but also upon the correlation amongst the low-fidelity models, and their relative computational cost. Because of this layer of complexity, the question of how to optimally select the set of low-fidelity models has remained open. In this work, methods for optimal model construction and tuning are investigated as a means to increase the speed and precision of trajectory simulation for EDL. Specifically, the focus is on the inclusion of low-fidelity model tuning within the sample allocation optimization that accompanies multi-model MC methods. Results indicate that low-fidelity model tuning can significantly improve efficiency and precision of trajectory simulations and provide an increased edge to multi-model MC methods when compared to standard MC.

Abstract not provided.

As transistors have been scaled over the past decade, modern systems have become increasingly susceptible to faults. Increased transistor densities and lower capacitances make a particle strike more likely to cause an upset. At the same time, complex computer systems are increasingly integrated into safety-critical systems such as autonomous vehicles. These two trends make the study of system reliability and fault tolerance essential for modern systems. To analyze and improve system reliability early in the design process, new tools are needed for RTL fault analysis.This paper proposes Eris, a novel framework to identify vulnerable components in hardware designs through fault-injection and fault propagation tracking. Eris builds on ESSENT - a fast C/C++ RTL simulation framework - to provide fault injection, fault tracking, and control-flow deviation detection capabilities for RTL designs. To demonstrate Eris' capabilities, we analyze the reliability of the open source Rocket Chip SoC by randomly injecting faults during thousands of runs on four microbenchmarks. As part of this analysis we measure the sensitivity of different hardware structures to faults based on the likelihood of a random fault causing silent data corruption, unrecoverable data errors, program crashes, and program hangs. We detect control flow deviations and determine whether or not they are benign. Additionally, using Eris' novel fault-tracking capabilities we are able to find 78% more vulnerable components in the same number of simulations compared to RTL-based fault injection techniques without these capabilities. We will release Eris as an open-source tool to aid future research into processor reliability and hardening.

This work evaluates the benefits of using a 'smart' network interface card (SmartNIC) as a compute accelerator for the example of the MiniMD molecular dynamics proxy application. The accelerator is NVIDIA's BlueField-2 card, which includes an 8-core Arm processor along with a small amount of DRAM and storage. We test the networking and data movement performance of these cards compared to a standard Intel server host using microbenchmarks and MiniMD. In MiniMD, we identify two distinct classes of computation, namely core computation and maintenance computation, which are executed in sequence. We restructure the algorithm and code to weaken this dependence and increase task parallelism, thereby making it possible to increase utilization of the BlueField-2 concurrently with the host. We evaluate our implementation on a cluster consisting of 16 dual-socket Intel Broadwell host nodes with one BlueField-2 per host-node. Our results show that while the overall compute performance of BlueField-2 is limited, using them with a modified MiniMD algorithm allows for up to 20% speedup over the host CPU baseline with no loss in simulation accuracy.

Fault tolerance is a key challenge as high performance computing systems continue to increase component counts, individual component reliability decreases, and hardware and software complexity increases. To better understand the potential impacts of failures on next-generation systems, significant effort has been devoted to collecting, characterizing and analyzing failures on current systems. These studies require large volumes of data and complex analysis in an attempt to identify statistical properties of the failure data. In this paper, we examine the lifetime of failures on the Cielo supercomputer that was located at Los Alamos National Laboratory, looking specifically at the time between faults on this system. Through this analysis, we show that the time between uncorrectable faults for this system obeys Benford’s law, This law applies to a number of naturally occurring collections of numbers and states that the leading digit is more likely to be small, for example a leading digit of 1 is more likely than 9. We also show that a number of common distributions used to model failures also follow this law. This work provides critical analysis on the distribution of times between failures for extreme-scale systems. Specifically, the analysis in this work could be used as a simple form of failure prediction or used for modeling realistic failures.

In recent years we have been exploring a novel asynchronous, ballistic physical model of reversible computing, variously termed ABRC (Asynchronous Ballistic Reversible Computing) or BARC (Ballistic Asynchronous Reversible Computing). In this model, localized information-bearing pulses propagate bidi-rectionally along nonbranching interconnects between I/O ports of stateful circuit elements, which carry out reversible transformations of the local digital state. The model appears suitable for implementation in superconducting circuits, using the naturally quantized configuration of magnetic flux in the circuit to encode digital information. One of the early research thrusts in this effort involves the enumeration and classification, at an abstract theoretical level, of the distinct possible reversible digital functional behaviors that primitive BARC circuit elements may exhibit, given the applicable conservation and symmetry constraints in superconducting implementations. In this paper, we describe the motivations for this work, outline our research methodology, and summarize some of the noteworthy preliminary results to date from our theoretical study of BARC elements for bipolarized pulses, and having up to three I/O ports and two internal digital states.

Abstract not provided.

We propose a new benchmark for high-performance (HP) computers. Similar to High Performance Conjugate Gradient (HPCG), the new benchmark is designed to rank computers based on how fast they can solve a sparse linear system of equations, exhibiting computational and communication requirements typical in many scientific applications. The main novelty of the new benchmark is that it is now based on Generalized Minimum Residual method (GMRES) (combined with Geometric Multi-Grid preconditioner and Gauss-Seidel smoother) and provides the flexibility to utilize lower precision arithmetic. This is motivated by new hardware architectures that deliver lower-precision arithmetic at higher performance. There are other machines that do not follow this trend. However, using a lower-precision arithmetic reduces the required amount of data transfer, which alone could improve solver performance. Considering these trends, an HP benchmark that allows the use of different precisions for solving important scientific problems will be valuable for many different disciplines, and we also hope to promote the design of future HP computers that can utilize mixed-precision arithmetic for achieving high application performance. We present our initial design of the new benchmark, its reference implementation, and the performance of the reference mixed (double and single) precision Geometric Multi-Grid solvers on current top-ranked architectures. We also discuss challenges of designing such a benchmark, along with our preliminary numerical results using 16-bit numerical values (half and bfloat precisions) for solving a sparse linear system of equations.

Abstract not provided.

Abstract not provided.

Abstract not provided.

Simple but mission-critical internet-based applications that require extremely high reliability, availability, and verifiability (e.g., auditability) could benefit from running on robust public programmable blockchain platforms such as Ethereum. Unfortunately, program code running on such blockchains is normally publicly viewable, rendering these platforms unsuitable for applications requiring strict privacy of application code, data, and results. In this work, we investigate using MPC techniques to protect the privacy of a blockchain computation. While our main goal is to hide both the data and the computed function itself, we also consider the standard MPC setting where the function is public. We describe GABLE (Garbled Autonomous Bots Leveraging Ethereum), a blockchain MPC architecture and system. The GABLE architecture specifies the roles and capabilities of the players. GABLE includes two approaches for implementing MPC over blockchain: Garbled Circuits (GC), evaluating universal circuits, and Garbled Finite State Automata (GFSA). We formally model and prove the security of GABLE implemented over garbling schemes, a popular abstraction of GC and GFSA from (Bellare et al., CCS 2012). We analyze in detail the performance (including Ethereum gas costs) of both approaches and discuss the trade-offs. We implement a simple prototype of GABLE and report on the implementation issues and experience.

To keep pace with the demand for innovation through scientific computing, modern scientific software development is increasingly reliant upon a rich and diverse ecosystem of software libraries and toolchains. Research software engineers (RSEs) responsible for that infrastructure perform highly integrative work, acting as a bridge between the hardware, the needs of researchers, and the software layers situated between them; relatively little, however, has been written about the role played by RSEs in that work and what support they need to thrive. To that end, we present a two-part report on the development of half-precision floating point support in the Kokkos Ecosystem. Half-precision computation is a promising strategy for increasing performance in numerical computing and is particularly attractive for emerging application areas (e.g., machine learning), but developing practicable, portable, and user-friendly abstractions is a nontrivial task. In the first half of the paper, we conduct an engineering study on the technical implementation of the Kokkos half-precision scalar feature and showcase experimental results; in the second half, we offer an experience report on the challenges and lessons learned during feature development by the first author. We hope our study provides a holistic view on scientific library development and surfaces opportunities for future studies into effective strategies for RSEs engaged in such work.

The s-step Conjugate Gradient (CG) algorithm has the potential to reduce the communication cost of standard CG by a factor of s. However, though mathematically equivalent, s-step CG may be numerically less stable compared to standard CG in finite precision, exhibiting slower convergence and decreased attainable accuracy. This limits the use of s-step CG in practice. To improve the numerical behavior of s-step CG and overcome this potential limitation, we incorporate two techniques. First, we improve convergence behavior through the use of higher precision at critical parts of the s-step iteration and second, we integrate a residual replacement strategy into the resulting mixed precision s-step CG to improve attainable accuracy. Our experimental results on the Summit Supercomputer demonstrate that when the higher precision is implemented in hardware, these techniques have virtually no overhead on the iteration time while improving both the convergence rate and the attainable accuracy of s-step CG. Even when the higher precision is implemented in software, these techniques may still reduce the time-to-solution (speedups of up to 1.8times in our experiments), especially when s-step CG suffers from numerical instability with a small step size and the latency cost becomes a significant part of its iteration time.

This paper reports on Catalyst usability and initial adoption by SPARC analysts. The use case approach highlights the analysts' perspective. Impediments to adoption can be due to deficiencies in software capabilities, or analysts may identify mundane inconveniences and barriers that prevent them from fully leveraging Catalyst. With that said, for many analyst tasks Catalyst provides enough relative advantage that they have begun applying it in their production work, and they recognize the potential for it to solve problems they currently struggle with. The findings in this report include specific issues and minor bugs in ParaView Python scripting, which are viewed as having straightforward solutions, as well as a broader adoption analysis.

We investigate the space complexity of two graph streaming problems: MAX-CUT and its quantum analogue, QUANTUM MAX-CUT. Previous work by Kapralov and Krachun [STOC 19] resolved the classical complexity of the classical problem, showing that any (2 - ?)-approximation requires O(n) space (a 2-approximation is trivial with O(log n) space). We generalize both of these qualifiers, demonstrating O(n) space lower bounds for (2 - ?)-approximating MAX-CUT and QUANTUM MAX-CUT, even if the algorithm is allowed to maintain a quantum state. As the trivial approximation algorithm for QUANTUM MAX-CUT only gives a 4-approximation, we show tightness with an algorithm that returns a (2 + ?)-approximation to the QUANTUM MAX-CUT value of a graph in O(log n) space. Our work resolves the quantum and classical approximability of quantum and classical Max-Cut using o(n) space.We prove our lower bounds through the techniques of Boolean Fourier analysis. We give the first application of these methods to sequential one-way quantum communication, in which each player receives a quantum message from the previous player, and can then perform arbitrary quantum operations on it before sending it to the next. To this end, we show how Fourier-analytic techniques may be used to understand the application of a quantum channel.

This manuscript presents the recent advances in Mixed-Integer Nonlinear Programming (MINLP) and Generalized Disjunctive Programming (GDP) with a particular scope for superstructure optimization within Process Systems Engineering (PSE). We present an environment of open-source software packages written in Python and based on the algebraic modeling language Pyomo. These packages include MindtPy, a solver for MINLP that implements decomposition algorithms for such problems, CORAMIN, a toolset for MINLP algorithms providing relaxation generators for nonlinear constraints, Pyomo.GDP, a modeling extension for Generalized Disjunctive Programming that allows users to represent their problem as a GDP natively, and GDPOpt, a collection of algorithms explicitly tailored for GDP problems. Combining these tools has allowed us to solve several problems relevant to PSE, which we have gathered in an easily installable and accessible library, GDPLib. We show two examples of these models and how the flexibility of modeling given by Pyomo.GDP allows for efficient solutions to these complex optimization problems. Finally, we show an example of integrating these tools with the framework IDAES PSE, leading to optimal process synthesis and conceptual design with advanced multi-scale PSE modeling systems.

Multi-model Monte Carlo methods have been illustrated to be an efficient and accurate alternative to standard Monte Carlo (MC) in the model-based propagation of uncertainty in entry, descent, and landing (EDL) applications. These multi-model MC methods fuse predictions from low-fidelity models with the high-fidelity EDL model of interest to produce unbiased statistics with a fraction of the computational cost. The accuracy and efficiency of the multi-model MC methods are dependent upon the magnitude of correlations of the low-fidelity models with the high-fidelity model, but also upon the correlation amongst the low-fidelity models, and their relative computational cost. Because of this layer of complexity, the question of how to optimally select the set of low-fidelity models has remained open. In this work, methods for optimal model construction and tuning are investigated as a means to increase the speed and precision of trajectory simulation for EDL. Specifically, the focus is on the inclusion of low-fidelity model tuning within the sample allocation optimization that accompanies multi-model MC methods. Results indicate that low-fidelity model tuning can significantly improve efficiency and precision of trajectory simulations and provide an increased edge to multi-model MC methods when compared to standard MC.

Numerical simulations of Greenland and Antarctic ice sheets involve the solution of large-scale highly nonlinear systems of equations on complex shallow geometries. This work is concerned with the construction of Schwarz preconditioners for the solution of the associated tangent problems, which are challenging for solvers mainly because of the strong anisotropy of the meshes and wildly changing boundary conditions that can lead to poorly constrained problems on large portions of the domain. Here, two-level generalized Dryja-Smith-Widlund (GDSW)-type Schwarz preconditioners are applied to different land ice problems, i.e., a velocity problem, a temperature problem, as well as the coupling of the former two problems. We employ the message passing interface (MPI)- parallel implementation of multilevel Schwarz preconditioners provided by the package FROSch (fast and robust Schwarz) from the Trilinos library. The strength of the proposed preconditioner is that it yields out-of-the-box scalable and robust preconditioners for the single physics problems. To the best of our knowledge, this is the first time two-level Schwarz preconditioners have been applied to the ice sheet problem and a scalable preconditioner has been used for the coupled problem. The preconditioner for the coupled problem differs from previous monolithic GDSW preconditioners in the sense that decoupled extension operators are used to compute the values in the interior of the subdomains. Several approaches for improving the performance, such as reuse strategies and shared memory OpenMP parallelization, are explored as well. In our numerical study we target both uniform meshes of varying resolution for the Antarctic ice sheet as well as nonuniform meshes for the Greenland ice sheet. We present several weak and strong scaling studies confirming the robustness of the approach and the parallel scalability of the FROSch implementation. Among the highlights of the numerical results are a weak scaling study for up to 32 K processor cores (8 K MPI ranks and 4 OpenMP threads) and 566 M degrees of freedom for the velocity problem as well as a strong scaling study for up to 4 K processor cores (and MPI ranks) and 68 M degrees of freedom for the coupled problem.

We have extended the computational singular perturbation (CSP) method to differential algebraic equation (DAE) systems and demonstrated its application in a heterogeneous-catalysis problem. The extended method obtains the CSP basis vectors for DAEs from a reduced Jacobian matrix that takes the algebraic constraints into account. We use a canonical problem in heterogeneous catalysis, the transient continuous stirred tank reactor (T-CSTR), for illustration. The T-CSTR problem is modelled fundamentally as an ordinary differential equation (ODE) system, but it can be transformed to a DAE system if one approximates typically fast surface processes using algebraic constraints for the surface species. We demonstrate the application of CSP analysis for both ODE and DAE constructions of a T-CSTR problem, illustrating the dynamical response of the system in each case. We also highlight the utility of the analysis in commenting on the quality of any particular DAE approximation built using the quasi-steady state approximation (QSSA), relative to the ODE reference case.

A primary objective of repository modeling is identification and assessment of features and processes providing safety performance. Sensitivity analyses typically provide information on how input parameters affect performance, not features and processes. To quantify the effects of features and processes, tracers can be introduced virtually in model simulations and tracked in informative ways. This paper describes five ways virtual tracers can be used to directly measure the relative importance of several features, processes, and combinations of features and processes in repository performance assessment modeling.

Centered on modern C++ and the SYCL standard for heterogeneous programming, Data Parallel C++ (dpc++) and Intel's oneAPI software ecosystem aim to lower the barrier to entry for the use of accelerators like FPGAs in diverse applications. In this work, we consider the usage of FPGAs for scientific computing, in particular with a multigrid solver, MueLu. We report on early experiences implementing kernels of the solver in DPC++ for execution on Stratix 10 FPGAs, and we evaluate several algorithmic design and implementation choices. These choices not only impact performance, but also shed light on the capabilities and limitations of DPC++ and oneAPI.

Abstract not provided.

Neural networks (NN)s have been increasingly proposed as surrogates for approximation of systems with computationally expensive physics for rapid online evaluation or exploration. As these surrogate models are integrated into larger optimization problems used for decision making, there is a need to verify their behavior to ensure adequate performance over the desired parameter space. We extend the ideas of optimization-based neural network verification to provide guarantees of surrogate performance over the feasible optimization space. In doing so, we present formulations to represent neural networks within decision-making problems, and we develop verification approaches that use model constraints to provide increasingly tight error estimates. We demonstrate the capabilities on a simple steady-state reactor design problem.

Stochasticity is ubiquitous in the world around us. However, our predominant computing paradigm is deterministic. Random number generation (RNG) can be a computationally inefficient operation in this system especially for larger workloads. Our work leverages the underlying physics of emerging devices to develop probabilistic neural circuits for RNGs from a given distribution. However, codesign for novel circuits and systems that leverage inherent device stochasticity is a hard problem. This is mostly due to the large design space and complexity of doing so. It requires concurrent input from multiple areas in the design stack from algorithms, architectures, circuits, to devices. In this paper, we present examples of optimal circuits developed leveraging AI-enhanced codesign techniques using constraints from emerging devices and algorithms. Our AI-enhanced codesign approach accelerated design and enabled interactions between experts from different areas of the micro-electronics design stack including theory, algorithms, circuits, and devices. We demonstrate optimal probabilistic neural circuits using magnetic tunnel junction and tunnel diode devices that generate an RNG from a given distribution.

Quantum computers can now run interesting programs, but each processor’s capability—the set of programs that it can run successfully—is limited by hardware errors. These errors can be complicated, making it difficult to accurately predict a processor’s capability. Benchmarks can be used to measure capability directly, but current benchmarks have limited flexibility and scale poorly to many-qubit processors. We show how to construct scalable, efficiently verifiable benchmarks based on any program by using a technique that we call circuit mirroring. With it, we construct two flexible, scalable volumetric benchmarks based on randomized and periodically ordered programs. We use these benchmarks to map out the capabilities of twelve publicly available processors, and to measure the impact of program structure on each one. We find that standard error metrics are poor predictors of whether a program will run successfully on today’s hardware, and that current processors vary widely in their sensitivity to program structure.

Centered on modern C++ and the SYCL standard for heterogeneous programming, Data Parallel C++ (dpc++) and Intel's oneAPI software ecosystem aim to lower the barrier to entry for the use of accelerators like FPGAs in diverse applications. In this work, we consider the usage of FPGAs for scientific computing, in particular with a multigrid solver, MueLu. We report on early experiences implementing kernels of the solver in DPC++ for execution on Stratix 10 FPGAs, and we evaluate several algorithmic design and implementation choices. These choices not only impact performance, but also shed light on the capabilities and limitations of DPC++ and oneAPI.

This paper describes an efficient reverse-mode differentiation algorithm for contraction operations for arbitrary and unconventional tensor network topologies. The approach leverages the tensor contraction tree of Evenbly and Pfeifer (2014), which provides an instruction set for the contraction sequence of a network. We show that this tree can be efficiently leveraged for differentiation of a full tensor network contraction using a recursive scheme that exploits (1) the bilinear property of contraction and (2) the property that trees have a single path from root to leaves. While differentiation of tensor-tensor contraction is already possible in most automatic differentiation packages, we show that exploiting these two additional properties in the specific context of contraction sequences can improve eficiency. Following a description of the algorithm and computational complexity analysis, we investigate its utility for gradient-based supervised learning for low-rank function recovery and for fitting real-world unstructured datasets. We demonstrate improved performance over alternating least-squares optimization approaches and the capability to handle heterogeneous and arbitrary tensor network formats. When compared to alternating minimization algorithms, we find that the gradient-based approach requires a smaller oversampling ratio (number of samples compared to number model parameters) for recovery. This increased efficiency extends to fitting unstructured data of varying dimensionality and when employing a variety of tensor network formats. Here, we show improved learning using the hierarchical Tucker method over the tensor-train in high-dimensional settings on a number of benchmark problems.

Many teams struggle to adapt and right-size software engineering best practices for quality assurance to fit their context. Introducing software quality is not usually framed in a way that motivates teams to take action, thus resulting in it becoming a "check the box for compliance"activity instead of a cultural practice that values software quality and the effort to achieve it. When and how can we provide effective incentives for software teams to adopt and integrate meaningful and enduring software quality practices? We explored this question through a persona-based ideation exercise at the 2021 Collegeville Workshop on Scientific Software in which we created three unique personas that represent different scientific software developer perspectives.

Using neural networks to solve variational problems, and other scientific machine learning tasks, has been limited by a lack of consistency and an inability to exactly integrate expressions involving neural network architectures. We address these limitations by formulating a polynomial-spline network, a novel shallow multilinear perceptron (MLP) architecture incorporating free knot B-spline basis functions into a polynomial mixture-of-experts model. Effectively, our architecture performs piecewise polynomial approximation on each cell of a trainable partition of unity while ensuring the MLP and its derivatives can be integrated exactly, obviating a reliance on sampling or quadrature and enabling error-free computation of variational forms. We demonstrate hp-convergence for regression problems at convergence rates expected from approximation theory and solve elliptic problems in one and two dimensions, with a favorable comparison to adaptive finite elements.

Given a graph, finding the distance-2 maximal independent set (MIS-2) of the vertices is a problem that is useful in several contexts such as algebraic multigrid coarsening or multilevel graph partitioning. Such multilevel methods rely on finding the independent vertices so they can be used as seeds for aggregation in a multilevel scheme. We present a parallel MIS-2 algorithm to improve performance on modern accelerator hardware. This algorithm is implemented using the Kokkos programming model to enable performance portability. We demonstrate the portability of the algorithm and the performance on a variety of architectures (x86/ARM CPUs and NVIDIA/AMD GPUs). The resulting algorithm is also deterministic, producing an identical result for a given input across all of these platforms. The new MIS-2 implementation outperforms implementations in state of the art libraries like CUSP and ViennaCL by 3-8x while producing similar quality results. We further demonstrate the benefits of this approach by developing parallel graph coarsening scheme for two different use cases. First, we develop an algebraic multigrid (AMG) aggregation scheme using parallel MIS-2 and demonstrate the benefits as opposed to previous approaches used in the MueLu multigrid package in Trilinos. We also describe an approach for implementing a parallel multicolor 'cluster' Gauss-Seidel preconditioner using this MIS-2 coarsening, and demonstrate better performance with an efficient, parallel, mul-ticolor Gauss-Seidel algorithm.

Powders under compression form mesostructures of particle agglomerations in response to both inter- and intra-particle forces. The ability to computationally predict the resulting mesostructures with reasonable accuracy requires models that capture the distributions associated with particle size and shape, contact forces, and mechanical response during deformation and fracture. The following report presents experimental data obtained for the purpose of validating emerging mesostructures simulated by discrete element method and peridynamic approaches. A custom compression apparatus, suitable for integration with our micro-computed tomography (micro-CT) system, was used to collect 3-D scans of a bulk powder at discrete steps of increasing compression. Details of the apparatus and the microcrystalline cellulose particles, with a nearly spherical shape and mean particle size, are presented. Comparative simulations were performed with an initial arrangement of particles and particle shapes directly extracted from the validation experiment. The experimental volumetric reconstruction was segmented to extract the relative positions and shapes of individual particles in the ensemble, including internal voids in the case of the microcrystalline cellulose particles. These computationally determined particles were then compressed within the computational domain and the evolving mesostructures compared directly to those in the validation experiment. The ability of the computational models to simulate the experimental mesostructures and particle behavior at increasing compression is discussed.

Stochasticity is ubiquitous in the world around us. However, our predominant computing paradigm is deterministic. Random number generation (RNG) can be a computationally inefficient operation in this system especially for larger workloads. Our work leverages the underlying physics of emerging devices to develop probabilistic neural circuits for RNGs from a given distribution. However, codesign for novel circuits and systems that leverage inherent device stochasticity is a hard problem. This is mostly due to the large design space and complexity of doing so. It requires concurrent input from multiple areas in the design stack from algorithms, architectures, circuits, to devices. In this paper, we present examples of optimal circuits developed leveraging AI-enhanced codesign techniques using constraints from emerging devices and algorithms. Our AI-enhanced codesign approach accelerated design and enabled interactions between experts from different areas of the micro-electronics design stack including theory, algorithms, circuits, and devices. We demonstrate optimal probabilistic neural circuits using magnetic tunnel junction and tunnel diode devices that generate an RNG from a given distribution.

Non-volatile memory arrays require select devices to ensure accurate programming. The one-selector one-resistor (1S1R) array where a two-terminal nonlinear select device is placed in series with a resistive memory element is attractive due to its high-density data storage; however, the effect of the nonlinear select device on the accuracy of analog in-memory computing has not been explored. This work evaluates the impact of select and memory device properties on the results of analog matrix-vector multiplications. We integrate nonlinear circuit simulations into CrossSim and perform end-to-end neural network inference simulations to study how the select device affects the accuracy of neural network inference. We propose an adjustment to the input voltage that can effectively compensate for the electrical load of the select device. Our results show that for deep residual networks trained on CIFAR-10, a compensation that is uniform across all devices in the system can mitigate these effects over a wide range of values for the select device I-V steepness and memory device On/Off ratio. A realistic I-V curve steepness of 60 mV/dec can yield an accuracy on CIFAR-10 that is within 0.44% of the floating-point accuracy.

The characterization of mixtures of non-interacting, spectroscopically similar quantum components has important applications in chemistry, biology, and materials science. We introduce an approach based on quantum tracking control that allows for determining the relative concentrations of constituents in a quantum mixture, using a single pulse which enhances the distinguishability of components of the mixture and has a length that scales linearly with the number of mixture constituents. To illustrate the method, we consider two very distinct model systems: mixtures of diatomic molecules in the gas phase, as well as solid-state materials composed of a mixture of components. A set of numerical analyses are presented, showing strong performance in both settings.