RAPIDS-2: a SciDAC Institute for Resource and Application Productivity through Computation, Information, and Data Science Final Report (University of Oregon)
Final report
Malony, Allen D. [University of Oregon] (ORCID:000↗
SEARCH · Engineering Papers
Results for “Computer science”
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.
Introducing Enabling Computational Tools to the Climate Sciences: Multi-Resolution Climate Modeling with Adaptive Cubed-Sphere Grids
The research investigates and advances strategies how to bridge the scale discrepancies between local, regional and global phenomena in climate models without the prohibitive computational costs of global cloud-resolving simulations. In particular, the research explores new frontiers in computational geoscience by introducing high-order Adaptive Mesh Refinement (AMR) techniques into climate research. AMR and statically-adapted variable-resolution approaches represent an emerging trend for atmospheric models and are likely to become the new norm in future-generation weather and climate models. The research advances the understanding of multi-scale interactions in the climate system and showcases a pathway how to model these interactions effectively with advanced computational tools, like the Chombo AMR library developed at the Lawrence Berkeley National Laboratory. The research is interdisciplinary and combines applied mathematics, scientific computing and the atmospheric sciences. In this research project, a hierarchy of high-order atmospheric models on cubed-sphere computational grids have been developed that serve as an algorithmic prototype for the finite-volume solution-adaptive Chombo-AMR approach. The foci of the investigations have lied on the characteristics of both static mesh adaptations and dynamically-adaptive grids that can capture flow fields of interest like tropical cyclones. Six research themes have been chosen. These are (1) the introduction of adaptive mesh refinement techniques into the climate sciences, (2) advanced algorithms for nonhydrostatic atmospheric dynamical cores, (3) an assessment of the interplay between resolved-scale dynamical motions and subgrid-scale physical parameterizations, (4) evaluation techniques for atmospheric model hierarchies, (5) the comparison of AMR refinement strategies and (6) tropical cyclone studies with a focus on multi-scale interactions and variable-resolution modeling. The results of this research project demonstrate significant advances in all six research areas. The major conclusions are that statically-adaptive variable-resolution modeling is currently becoming mature in the climate sciences, and that AMR holds outstanding promise for future-generation weather and climate models on high-performance computing architectures.
58 GEOSCIENCES↗
What is the current state of the science of Cyber defense?
My overall sense of the cyber defense field is one of an adolescent discipline currently bogged down in a cloud of issues, the most iconic of which is the great diversity of approaches that are being aggregated to form a coherent field. Because my own expertise is complex systems and materials physics research, I have limited direct experience in cyber security sciences except as a user of secure networks and computing resources. However, in producing this report, I have found with certainty that there exists no calculus for cyber risk assessment, mitigation, and response, although some hopeful precepts toward this end are emerging.
Computer Science↗
Exascale Co-design for Modeling Materials in Extreme Environments
Computational materials science has provided great insight into the response of materials under extreme conditions that are difficult to probe experimentally. For example, shock-induced plasticity and phase transformation processes in single-crystal and nanocrystalline metals have been widely studied via large-scale molecular dynamics simulations, and many of these predictions are beginning to be tested at advanced 4th generation light sources such as the Advanced Photon Source (APS) and Linac Coherent Light Source (LCLS). I will describe our simulation predictions and their recent verification at LCLS, outstanding challenges in modeling the response of materials to extreme mechanical and radiation environments, and our efforts to tackle these as part of the multi-institutional, multi-disciplinary Exascale Co-design Center for Materials in Extreme Environments (ExMatEx). ExMatEx has initiated an early and deep collaboration between domain (computational materials) scientists, applied mathematicians, computer scientists, and hardware architects, in order to establish the relationships between algorithms, software stacks, and architectures needed to enable exascale-ready materials science application codes within the next decade. We anticipate that we will be able to exploit hierarchical, heterogeneous architectures to achieve more realistic large-scale simulations with adaptive physics refinement, and are using tractable application scale-bridging proxy application testbeds to assess new approaches and requirements. Such current scale-bridging strategies accumulate (or recompute) a distributed response database from fine-scale calculations, in a top-down rather than bottom-up multiscale approach.
36 MATERIALS SCIENCE↗
Large Scale Computing and Storage Requirements for Fusion Energy Sciences: Target 2017
The National Energy Research Scientific Computing Center (NERSC) is the primary computing center for the DOE Office of Science, serving approximately 4,500 users working on some 650 projects that involve nearly 600 codes in a wide variety of scientific disciplines. In March 2013, NERSC, DOE's Office of Advanced Scientific Computing Research (ASCR) and DOE's Office of Fusion Energy Sciences (FES) held a review to characterize High Performance Computing (HPC) and storage requirements for FES research through 2017. This report is the result.
70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
Computing rank‐revealing factorizations of matrices stored out‐of‐core
This paper describes efficient algorithms for computing rank-revealing factorizations of matrices that are too large to fit in main memory (RAM), and must instead be stored on slow external memory devices such as disks (out-of-core or out-of-memory). Traditional algorithms for computing rank-revealing factorizations (such as the column pivoted QR factorization and the singular value decomposition) are very communication intensive as they require many vector-vector and matrix-vector operations, which become prohibitively expensive when data is not in RAM. Randomization allows to reformulate new methods so that large contiguous blocks of the matrix are processed in bulk. The paper describes two distinct methods. The first is a blocked version of column pivoted Householder QR, organized as a “left-looking” method to minimize the number of the expensive write operations. The second method results employs a UTV factorization. It is organized as an algorithm-by-blocks to overlap computations and I/O operations. As it incorporates power iterations, it is much better at revealing the numerical rank. Numerical experiments on several computers demonstrate that the new algorithms are almost as fast when processing data stored on slow memory devices as traditional algorithms are for data stored in RAM.
97 MATHEMATICS AND COMPUTING↗
Fast GPU 3D diffeomorphic image registration
3D image registration is one of the most fundamental and computationally expensive operations in medical image analysis. Here, we present a mixed-precision, Gauss–Newton–Krylov solver for diffeomorphic registration of two images. Our work extends the publicly available CLAIRE library to GPU architectures. Despite the importance of image registration, only a few implementations of large deformation diffeomorphic registration packages support GPUs. Our contributions are new algorithms to significantly reduce the run time of the two main computational kernels in CLAIRE: calculation of derivatives and scattered-data interpolation. Additionally, we deploy (i) highly-optimized, mixed-precision GPU-kernels for the evaluation of scattered-data interpolation, (ii) replace Fast-Fourier-Transform (FFT)-based first-order derivatives with optimized 8th-order finite differences, and (iii) compare with state-of-the-art CPU and GPU implementations. As a highlight, we demonstrate that we can register clinical images in less than 6 s on a single NVIDIA Tesla V100. This amounts to over 20 speed-up over the current version of CLAIRE and over 30 speed-up over existing GPU implementations.
97 MATHEMATICS AND COMPUTING↗
Trade-offs between individual and ensemble forecasts of an emerging infectious disease
Probabilistic forecasts play an indispensable role in answering questions about the spread of newly emerged pathogens. However, uncertainties about the epidemiology of emerging pathogens can make it difficult to choose among alternative model structures and assumptions. To assess the potential for uncertainties about emerging pathogens to affect forecasts of their spread, we evaluated the performance 16 forecasting models in the context of the 2015-2016 Zika epidemic in Colombia. Each model featured a different combination of assumptions about human mobility, spatiotemporal variation in transmission potential, and the number of virus introductions. We found that which model assumptions had the most ensemble weight changed through time. We additionally identified a trade-off whereby some individual models outperformed ensemble models early in the epidemic, but on average the ensembles outperformed all individual models. Our results suggest that multiple models spanning uncertainty across alternative assumptions are necessary to obtain robust forecasts for emerging infectious diseases.
60 APPLIED LIFE SCIENCES↗
Optimization of nutrient utilization efficiency and productivity for algal cultures under light and dark cycles using genome-scale model process control
Abstract Algal cultivations are strongly influenced by light and dark cycles. In this study, genome-scale metabolic models were applied to optimize nutrient supply during alternating light and dark cycles of Chlorella vulgaris . This approach lowered the glucose requirement by 75% and nitrate requirement by 23%, respectively, while maintaining high final biomass densities that were more than 80% of glucose-fed heterotrophic culture. Furthermore, by strictly controlling glucose feeding during the alternating cycles based on model-input, yields of biomass, lutein, and fatty acids per gram of glucose were more than threefold higher with cycling compared to heterotrophic cultivation. Next, the model was incorporated into open-loop and closed-loop control systems and compared with traditional fed-batch systems. Closed-loop systems which incorporated a feed-optimizing algorithm increased biomass yield on glucose more than twofold compared to standard fed-batch cultures for cycling cultures. Finally, the performance was compared to conventional proportional-integral-derivative (PID) controllers. Both simulation and experimental results exhibited superior performance for genome-scale model process control (GMPC) compared to traditional PID systems, reducing the overall measured value and setpoint error by 80% over 8 h. Overall, this approach provides researchers with the capability to enhance nutrient utilization and productivity of cell factories systematically by combining genome-scale models and controllers into an integrated platform with superior performance to conventional fed-batch and PID methodologies.
59 BASIC BIOLOGICAL SCIENCES↗
Real-time mixed reality display of dual particle radiation detector data
Radiation source localization and characterization are challenging tasks that currently require complex analyses for interpretation. Mixed reality (MR) technologies are at the verge of wide scale adoption and can assist in the visualization of complex data. Herein, we demonstrate real-time visualization of gamma ray and neutron radiation detector data in MR using the Microsoft HoloLens 2 smart glasses, significantly reducing user interpretation burden. Radiation imaging systems typically use double-scatter events of gamma rays or fast neutrons to reconstruct the incidence directional information, thus enabling source localization. The calculated images and estimated ’hot spots’ are then often displayed in 2D angular space projections on screens. By combining a state-of-the-art dual particle imaging system with HoloLens 2, we propose to display the data directly to the user via the head-mounted MR smart glasses, presenting the directional information as an overlay to the user’s 3D visual experience. We describe an open source implementation using efficient data transfer, image calculation, and 3D engine. We thereby demonstrate for the first time a real-time user experience to display fast neutron or gamma ray images from various radioactive sources set around the detector. We also introduce an alternative source search mode for situations of low event rates using a neural network and simulation based training data to provide a fast estimation of the source’s angular direction. Using MR for radiation detection provides a more intuitive perception of radioactivity and can be applied in routine radiation monitoring, education & training, emergency scenarios, or inspections.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
Pan- and core- network analysis of co-expression genes in a model plant
Genome-wide gene expression experiments have been performed using the model plant Arabidopsis during the last decade. Some studies involved construction of coexpression networks, a popular technique used to identify groups of co-regulated genes, to infer unknown gene functions. One approach is to construct a single coexpression network by combining multiple expression datasets generated in different labs. We advocate a complementary approach in which we construct a large collection of 134 coexpression networks based on expression datasets reported in individual publications. To this end we reanalyzed public expression data. To describe this collection of networks we introduced concepts of ‘pan-network’ and ‘core-network’ representing union and intersection between a sizeable fractions of individual networks, respectively. Here, we showed that these two types of networks are different both in terms of their topology and biological function of interacting genes. For example, the modules of the pan-network are enriched in regulatory and signaling functions, while the modules of the core-network tend to include components of large macromolecular complexes such as ribosomes and photosynthetic machinery. Our analysis is aimed to help the plant research community to better explore the information contained within the existing vast collection of gene expression data in Arabidopsis.
60 APPLIED LIFE SCIENCES↗
Dual-rail encoding with superconducting cavities
The design of quantum hardware that reduces and mitigates errors is essential for practical quantum error correction (QEC) and useful quantum computation. To this end, we introduce the circuit-Quantum Electrodynamics (QED) dual-rail qubit in which our physical qubit is encoded in the single-photon subspace, { | 01 〉 , | 10 〉 } , of two superconducting microwave cavities. The dominant photon loss errors can be detected and converted into erasure errors, which are in general much easier to correct. In contrast to linear optics, a circuit-QED implementation of the dual-rail code offers unique capabilities. Using just one additional transmon ancilla per dual-rail qubit, we describe how to perform a gate-based set of universal operations that includes state preparation, logical readout, and parametrizable single and two-qubit gates. Moreover, first-order hardware errors in the cavities and the transmon can be detected and converted to erasure errors in all operations, leaving background Pauli errors that are orders of magnitude smaller. Hence, the dual-rail cavity qubit exhibits a favorable hierarchy of error rates and is expected to perform well below the relevant QEC thresholds with today’s coherence times.
97 MATHEMATICS AND COMPUTING↗
Quantum eigenvalue estimation via time series analysis
We present an efficient method for estimating the eigenvalues of a Hamiltonian H from the expectation values of the evolution operator for various times. For a given quantum state ρ, our method outputs a list of eigenvalue estimates and approximate probabilities. Each probability depends on the support of ρ in those eigenstates of H associated with eigenvalues within an arbitrarily small range. The complexity of our method is polynomial in the inverse of a given precision parameter ε, which is the gap between eigenvalue estimates. Unlike the well-known quantum phase estimation algorithm that uses the quantum Fourier transform, our method does not require large ancillary systems, large sequences of controlled operations, or preserving coherence between experiments, and is therefore more attractive for near-term applications. The output of our method can be used to estimate spectral properties of H and other expectation values efficiently, within additive error proportional to ε.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
A review of non-cognitive applications for neuromorphic computing
Abstract Though neuromorphic computers have typically targeted applications in machine learning and neuroscience (‘cognitive’ applications), they have many computational characteristics that are attractive for a wide variety of computational problems. In this work, we review the current state-of-the-art for non-cognitive applications on neuromorphic computers, including simple computational kernels for composition, graph algorithms, constrained optimization, and signal processing. We discuss the advantages of using neuromorphic computers for these different applications, as well as the challenges that still remain. The ultimate goal of this work is to bring awareness to this class of problems for neuromorphic systems to the broader community, particularly to encourage further work in this area and to make sure that these applications are considered in the design of future neuromorphic systems.
97 MATHEMATICS AND COMPUTING↗
Reconfigurable Framework for Resilient Semantic Segmentation for Space Applications
Deep learning (DL) presents new opportunities for enabling spacecraft autonomy, onboard analysis, and intelligent applications for space missions. However, DL applications are computationally intensive and often infeasible to deploy on radiation-hardened (rad-hard) processors, which traditionally harness a fraction of the computational capability of their commercial-off-the-shelf counterparts. Commercial FPGAs and system-on-chips present numerous architectural advantages and provide the computation capabilities to enable onboard DL applications; however, these devices are highly susceptible to radiation-induced single-event effects (SEEs) that can degrade the dependability of DL applications. In this article, we propose Reconfigurable ConvNet (RECON), a reconfigurable acceleration framework for dependable, high-performance semantic segmentation for space applications. In RECON, we propose both selective and adaptive approaches to enable efficient SEE mitigation. In our selective approach, control-flow parts are selectively protected by triple-modular redundancy to minimize SEE-induced hangs, and in our adaptive approach, partial reconfiguration is used to adapt the mitigation of dataflow parts in response to a dynamic radiation environment. Combined, both approaches enable RECON to maximize system performability subject to mission availability constraints. We perform fault injection and neutron irradiation to observe the susceptibility of RECON and use dependability modeling to evaluate RECON in various orbital case studies to demonstrate a 1.5–3.0× performability improvement in both performance and energy efficiency compared to static approaches.
97 MATHEMATICS AND COMPUTING↗
If We Build Them, They Will Run: Automated HPC Apps Deployment and Profiling with eBPF in Cloud
The high performance computing (HPC) community is in a period of transition. The rise of AI/ML coupled with a changing landscape of resources deems portability a new metric of performance, and methods to move between on-premises and cloud environments and assess compatibility are paramount. Here we design and test a strategy for bridging the gap between traditional HPC and Kubernetes environments – first containerizing applications, providing automated orchestration to run studies, and packaging the setup with automated means to assess performance using low overhead eXtended Berkeley Packet Filter (eBPF) programs. We first assess different designs for eBPF collection, demonstrating a tradeoff between number of programs deployed on a node and overhead added. We develop 5 low overhead eBPF programs that combine with streaming ML models to assess CPU, futex, TCP, shared memory, and file access across four different builds of an HPC application for CPU and GPU. We use eBPF data to generate insights into the possible underlying etiology of scaling issues. We then assess compatibility of a well-known benchmark, HPCG, across matrices of micro-architectures and optimization levels (217 containers across 24 instance types and over 7500 runs). We provide to the community 30 applications to deploy in our automated setup and perform a scaling study from 4 to a maximum of 256 nodes for both CPU and GPU applications. Finally, we use our gained knowledge about performance to generate compatibility artifacts that are used by a newly developed Kubernetes controller to intelligently select instance type based on optimizing a figure of merit. Along with insights to scaling in this environment with a collection of applications and templates to work from, we provide an overall strategy for approaching HPC application deployment and image selection based on compatibility in cloud.
Computer science↗
Laboratory Directed Research and Development Annual Report for 2012
This report documents progress made on all LDRD-funded projects during fiscal year 2012.
Advanced Sensors and Instrumentation↗
Laboratory Directed Research and Development Annual Report (FY 2013)
The mission of Pacific Northwest National Laboratory is to transform the world through courageous discovery and innovation. Every day, scientists and engineers at PNNL set out to fulfill the promise of that mission by advancing scientific discovery and delivering critical solutions to the challenges that face our nation and the world. The science and technology we work with, create, and improve upon inspires and enables the world to live prosperously, safely, and securely. Our investments in Laboratory Directed Research and Development are essential to our ability to advance our strategy and realize this mission. LDRD investments help us nurture science and technology capabilities while capitalizing on the breadth and depth of talent that our staff possess. This report describes how we conduct our LDRD program in compliance with DOE objectives and guidelines. We use rigorous internal and external peer review to maintain the scientific value and soundness of the research enabled by our LDRD program.
advanced sensors and instrumentation↗