Search NASA⌕ Search

SEARCH · Search NASA

Results for “graph learning”

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 847 records · Page 47

Performance optimization, modeling and analysis of sparse matrix-matrix products on multi-core and many-core processors

Sparse matrix-matrix multiplication (SpGEMM) is a computational primitive that is widely used in areas ranging from traditional numerical applications to recent big data analysis and machine learning. Although many SpGEMM algorithms have been proposed, hardware specific optimizations for multi- and many-core processors are lacking and a detailed analysis of their performance under various use cases and matrices is not available. In this work, we firstly identify and mitigate multiple bottlenecks with memory management and thread scheduling on Intel Xeon Phi (Knights Landing or KNL). Specifically targeting many-core processors, we develop a hash-table-based algorithm and optimize a heap-based shared-memory SpGEMM algorithm. We examine their performance together with other publicly available codes. Different from the literature, our evaluation also includes use cases that are representative of real graph algorithms, such as multi-source breadth-first search or triangle counting. Our hash-table and heap-based algorithms are showing significant speedups from libraries in the majority of the cases while different algorithms dominate the other scenarios with different matrix size, sparsity, compression factor and operation type. We wrap up in-depth evaluation results and make a recipe to give the best SpGEMM algorithm for target scenario. We build the performance model for hash-table and heap-based algorithms, which supports the recipe. A critical finding is that hash-table-based SpGEMM gets a significant performance boost if the nonzeros are not required to be sorted within each row of the output matrix. Finally, we integrate our implementations into a large-scale protein clustering code named HipMCL, accelerating its SpGEMM kernel by up to 10X and achieving an overall performance boost for the whole HipMCL application by 2.6X.

97 MATHEMATICS AND COMPUTING↗

Jet Noise Reduction

The Acoustics Branch is responsible for reducing noise levels for jet and fan components on aircraft engines. To do this, data must be measured and calibrated accurately to ensure validity of test results. This noise reduction is accomplished by modifications to hardware such as jet nozzles, and by the use of other experimental hardware such as fluidic chevrons, elliptic cores, and fluidic shields. To insure validity of data calibration, a variety of software is used. This software adjusts the sound amplitude and frequency to be consistent with data taken on another day. Both the software and the hardware help make noise reduction possible. work properly. These software programs were designed to make corrections for atmosphere, shear, attenuation, electronic, and background noise. All data can be converted to a one-foot lossless condition, using the proper software corrections, making a reading independent of weather and distance. Also, data can be transformed from model scale to full scale for noise predictions of a real flight. Other programs included calculations of Over All Sound Pressure Level (OASPL), Effective Perceived Noise Level (EPNL). OASPL is the integration of sound with respect to frequency, and EPNL is weighted for a human s response to different sound frequencies and integrated with respect to time. With the proper software correction, data taken in the NATR are useful in determining ways to reduce noise. display any difference between two or more data files. Using this program and graphs of the data, the actual and predicted data can be compared. This software was tested on data collected at the Aero Acoustic Propulsion Laboratory (AAPL) using a variety of window types and overlaps. Similarly, short scripts were written to test each individual program in the software suite for verification. Each graph displays both the original points and the adjusted points connected with lines. During this summer, data points were taken during a live experiment at the AAPL to measure Nozzle Acoustic Test Rig (NATR) background noise levels. Six condenser microphones were placed in strategic locations around the dome and the inlet tunnel to measure different noise sources. From the control room the jet was monitored with the help of video cameras and other sensors. The data points were recorded, reduced, and plotted, and will be used to plan future modifications to the NATR. The primary goal to create data reduction test programs and provide verification was completed. As a result of the internship, I learned C/C++, UNIX/LINUX, Excel, and acoustic data processing methods. I also recorded data at the AAPL, then processed and plotted it. These data would be useful to compare against existing data. In addition, I adjusted software to work on the Mac OSX platform. And I used the available training resources.

Kenny, Patrick↗

Understanding the Scalability of Bayesian Network Inference using Clique Tree Growth Curves

Bayesian networks (BNs) are used to represent and efficiently compute with multi-variate probability distributions in a wide range of disciplines. One of the main approaches to perform computation in BNs is clique tree clustering and propagation. In this approach, BN computation consists of propagation in a clique tree compiled from a Bayesian network. There is a lack of understanding of how clique tree computation time, and BN computation time in more general, depends on variations in BN size and structure. On the one hand, complexity results tell us that many interesting BN queries are NP-hard or worse to answer, and it is not hard to find application BNs where the clique tree approach in practice cannot be used. On the other hand, it is well-known that tree-structured BNs can be used to answer probabilistic queries in polynomial time. In this article, we develop an approach to characterizing clique tree growth as a function of parameters that can be computed in polynomial time from BNs, specifically: (i) the ratio of the number of a BN's non-root nodes to the number of root nodes, or (ii) the expected number of moral edges in their moral graphs. Our approach is based on combining analytical and experimental results. Analytically, we partition the set of cliques in a clique tree into different sets, and introduce a growth curve for each set. For the special case of bipartite BNs, we consequently have two growth curves, a mixed clique growth curve and a root clique growth curve. In experiments, we systematically increase the degree of the root nodes in bipartite Bayesian networks, and find that root clique growth is well-approximated by Gompertz growth curves. It is believed that this research improves the understanding of the scaling behavior of clique tree clustering, provides a foundation for benchmarking and developing improved BN inference and machine learning algorithms, and presents an aid for analytical trade-off studies of clique tree clustering using growth curves.

Mengshoel, Ole Jakob↗

MICCO: An Enhanced Multi-GPU Scheduling Framework for Many-Body Correlation Functions

Calculation of many-body correlation functions is one of the critical kernels utilized in many scientific computing areas, especially in Lattice Quantum Chromodynamics (Lattice QCD). It is formalized as a sum of a large number of contraction terms each of which can be represented by a graph consisting of vertices describing quarks inside a hadron node and edges designating quark propagations at specific time intervals. Due to its computation- and memory-intensive nature, real-world physics systems (e.g., multi-meson or multi-baryon systems) explored by Lattice QCD prefer to leverage multi-GPUs. Different from general graph processing, many-body correlation function calculations show two specific features: a large number of computation-/data-intensive kernels and frequently repeated appearances of original and intermediate data. The former results in expensive memory operations such as tensor movements and evictions. The latter offers data reuse opportunities to mitigate the data-intensive nature of many-body correlation function calculations. However, existing graph-based multi-GPU schedulers cannot capture these data-centric features, thus resulting in a sub-optimal performance for many-body correlation function calculations. To address this issue, this paper presents a multi-GPU scheduling framework, MICCO, to accelerate contractions for correlation functions particularly by taking the data dimension (e.g., data reuse and data eviction) into account. This work first performs a comprehensive study on the interplay of data reuse and load balance, and designs two new concepts: local reuse pattern and reuse bound to study the opportunity of achieving the optimal trade-off between them. Based on this study, MICCO proposes a heuristic scheduling algorithm and a machine-learning-based regression model to generate the optimal setting of reuse bounds. Specifically, MICCO is integrated into a real-world Lattice QCD system, Redstar, for the first time running on multiple GPUs. The evaluation demonstrates MICCO outperforms other state-of-art works, achieving up to 2.25× speedup in synthesized datasets, and 1.49× speedup in real-world correlation functions.

Wang, Qihan↗

GAHLS: an optimized graph analytics based high level synthesis framework

The urgent need for low latency, high-compute and low power on-board intelligence in autonomous systems, cyber-physical systems, robotics, edge computing, evolvable computing, and complex data science calls for determining the optimal amount and type of specialized hardware together with reconfigurability capabilities. With these goals in mind, we propose a novel comprehensive graph analytics based high level synthesis (GAHLS) framework that efficiently analyzes complex high level programs through a combined compiler-based approach and graph theoretic optimization and synthesizes them into message passing domain-specific accelerators. This GAHLS framework first constructs a compiler-assisted dependency graph (CaDG) from low level virtual machine (LLVM) intermediate representation (IR) of high level programs and converts it into a hardware friendly description representation. Next, the GAHLS framework performs a memory design space exploration while account for the identified computational properties from the CaDG and optimizing the system performance for higher bandwidth. The GAHLS framework also performs a robust optimization to identify the CaDG subgraphs with similar computational structures and aggregate them into intelligent processing clusters in order to optimize the usage of underlying hardware resources. Finally, the GAHLS framework synthesizes this compressed specialized CaDG into processing elements while optimizing the system performance and area metrics. Evaluations of the GAHLS framework on several real-life applications (e.g., deep learning, brain machine interfaces) demonstrate that it provides 14.27× performance improvements compared to state-of-the-art approaches such as LegUp 6.2.

97 MATHEMATICS AND COMPUTING↗

Scalable Topological Data Analysis and Visualization for Evaluating Data-Driven Models in Scientific Applications

With the rapid adoption of machine learning techniques for large-scale applications in science and engineering comes the convergence of two grand challenges in visualization. First, the utilization of black box models (e.g., deep neural networks) calls for advanced techniques in exploring and interpreting model behaviors. Second, the rapid growth in computing has produced enormous datasets that require techniques that can handle millions or more samples. Although some solutions to these interpretability challenges have been proposed, they typically do not scale beyond thousands of samples, nor do they provide the high-level intuition scientists are looking for. Here, we present the first scalable solution to explore and analyze high-dimensional functions often encountered in the scientific data analysis pipeline. By combining a new streaming neighborhood graph construction, the corresponding topology computation, and a novel data aggregation scheme, namely topology aware datacubes, we enable interactive exploration of both the topological and the geometric aspect of high-dimensional data. Following two use cases from high-energy-density (HED) physics and computational biology, we demonstrate how these capabilities have led to crucial new insights in both applications.

97 MATHEMATICS AND COMPUTING↗

Population Simulation, AKA: Grahz, Rahbitz and Fawkzes

In an effort to give students a more visceral experience of science and instill a deeper working knowledge of concepts, activities that utilize hands-on, laboratory and simulated experiences are recommended because these activities have a greater impact on student learning, especially for Native American students. Because it is not usually feasible to take large and/or multiple classes of high school science students into the field to count numbers of organisms of a particular species, especially over a long period of time and covering a large area of an environment, the population simulation presented in this paper was created to aid students in understanding population dynamics by working with a simulated environment, which can be done in the classroom. Students create an environment and populate the environment with imaginary species. Then, using a sequence of "rules" that allow organisms to eat, reproduce, move and age, students see how the population of a species changes over time. In particular, students practice collecting data, summarizing information, plotting graphs, and interpreting graphs for such information as carrying capacity, predator prey relationships, and how specific species factors impact population and the environment. Students draw conclusions from their results and suggest further research, which may involve changes in simulation parameters, prediction of outcomes, and testing predictions. The population Simulation has demonstrated success in the above student activities using a "board game" version of the population simulation. A computer version of the population simulation needs more testing, but preliminary runs are promising. A second - and more complicated - computer simulation will simulate the same things and will add simulated population genetics.

Bangert, Tyler R.↗

Lorentz group equivariant autoencoders

Abstract There has been significant work recently in developing machine learning (ML) models in high energy physics (HEP) for tasks such as classification, simulation, and anomaly detection. Often these models are adapted from those designed for datasets in computer vision or natural language processing, which lack inductive biases suited to HEP data, such as equivariance to its inherent symmetries. Such biases have been shown to make models more performant and interpretable, and reduce the amount of training data needed. To that end, we develop the Lorentz group autoencoder (LGAE), an autoencoder model equivariant with respect to the proper, orthochronous Lorentz group $$\textrm{SO}^+(3,1)$$ SO + ( 3 , 1 ) , with a latent space living in the representations of the group. We present our architecture and several experimental results on jets at the LHC and find it outperforms graph and convolutional neural network baseline models on several compression, reconstruction, and anomaly detection metrics. We also demonstrate the advantage of such an equivariant model in analyzing the latent space of the autoencoder, which can improve the explainability of potential anomalies discovered by such ML models.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Sparse Convolutional Neural Networks for particle classification in ProtoDUNE-SP events

Deep Learning (DL) methods and Computer Vision are becoming important tools for event reconstruction in particle physics detectors. In this work, we report on the use of submanifold sparse convolutional neural networks (SparseNets) for the classification of track and shower hits from a DUNE prototype liquid-argon detector at CERN (ProtoDUNE-SP). By taking advantage of the three-dimensional nature of the problem we use a set of nine input features to classify sparse and locally dense hits associated to track or shower particles. The SparseNet has been trained on a test sample and shows promising results: efficiencies and purities greater than 90%. This has also been achieved with a considerable speedup and substantially less resource utilization with respect to other DL networks such as graph neural networks. This method offers great scalability advantages for future large neutrino detectors such as the planned DUNE experiment.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA Tech Briefs, September 2013

Topics include: ISS Ammonia Leak Detection Through X-Ray Fluorescence; A System for Measuring the Sway of the Vehicle Assembly Building; Fast, High-Precision Readout Circuit for Detector Arrays; Victim Simulator for Victim Detection Radar; Hydrometeor Size Distribution Measurements by Imaging the Attenuation of a Laser Spot; Quasi-Linear Circuit; High-Speed, High-Resolution Time-to-Digital Conversion; Li-Ion Battery and Supercapacitor Hybrid Design for Long Extravehicular Activities; Ultrasonic Low-Friction Containment Plate for Thermal and Ultrasonic Stir Weld Processes; High-Powered, Ultrasonically Assisted Thermal Stir Welding; Next-Generation MKIII Lightweight HUT/Hatch Assembly; Centrifugal Sieve for Gravity-Level-Independent Size; Segregation of Granular Materials; Ion Exchange Technology Development in Support of the Urine Processor Assembly; Nickel-Graphite Composite Compliant Interface and/or Hot Shoe Material; UltraSail CubeSat Solar Sail Flight Experiment; Mechanism for Deploying a Long, Thin-Film Antenna From a Rover; Counterflow Regolith Heat Exchanger; Acquisition and Retaining Granular Samples via a Rotating Coring Bit; Very-Low-Cost, Rugged Vacuum System; Medicine Delivery Device With Integrated Sterilization and Detection; FRET-Aptamer Assays for Bone Marker Assessment, C-Telopeptide, Creatinine, and Vitamin D; Multimode Directional Coupler for Utilization of Harmonic Frequencies from TWTAs; Dual-Polarization, Multi-Frequency Antenna Array for use with Hurricane Imaging Radiometer; Complementary Barrier Infrared Detector (CBIRD) Contact Methods; Autonomous Control of Space Nuclear Reactors; High-Power, High-Speed Electro-Optic Pockels Cell Modulator; Covariance Analysis Tool (G-CAT) for Computing Ascent, Descent, and Landing Errors; Enigma Version 12; Micrometeoroid and Orbital Debris (MMOD) Shield Ballistic Limit Analysis Program; Spitzer Telemetry Processing System; Planetary Protection Bioburden Analysis Program; Wing Leading Edge RCC Rapid Response Damage Prediction Tool (IMPACT2); ISSM: Ice Sheet System Model; Automated Loads Analysis System (ATLAS); Integrated Main Propulsion System Performance Reconstruction Process/Models. Phoenix Telemetry Processor; Contact Graph Routing Enhancements Developed in ION for DTN; GFEChutes Lo-Fi; Advanced Strategic and Tactical Relay Request Management for the Mars Relay Operations Service; Software for Generating Troposphere Corrections for InSAR Using GPS and Weather Model Data; Ionospheric Specifications for SAR Interferometry (ISSI); Implementation of a Wavefront-Sensing Algorithm; Sally Ride EarthKAM - Automated Image Geo-Referencing Using Google Earth Web Plug-In; Trade Space Specification Tool (TSST) for Rapid Mission Architecture (Version 1.2); Acoustic Emission Analysis Applet (AEAA) Software; Memory-Efficient Onboard Rock Segmentation; Advanced Multimission Operations System (ATMO); Robot Sequencing and Visualization Program (RSVP); Automating Hyperspectral Data for Rapid Response in Volcanic Emergencies; Raster-Based Approach to Solar Pressure Modeling; Space Images for NASA JPL Android Version; Kinect Engineering with Learning (KEWL); Spacecraft 3D Augmented Reality Mobile App; MPST Software: grl_pef_check; Real-Time Multimission Event Notification System for Mars Relay; SIM_EXPLORE: Software for Directed Exploration of Complex Systems; Mobile Timekeeping Application Built on Reverse-Engineered JPL Infrastructure; Advanced Query and Data Mining Capabilities for MaROS; Jettison Engineering Trajectory Tool; MPST Software: grl_suppdoc; PredGuid+A: Orion Entry Guidance Modified for Aerocapture; Planning Coverage Campaigns for Mission Design and Analysis: CLASP for DESDynl; and Space Place Prime.

Source record↗

Molecular Assemblies, Genes and Genomics Integrated Efficiently (MAGGIE)

Final report on MAGGIE. We set ambitious goals to model the functions of individual organisms and their community from molecular to systems scale. These scientific goals are driving the development of sophisticated algorithms to analyze large amounts of experimental measurements made using high throughput technologies to explain and predict how the environment influences biological function at multiple scales and how the microbial systems in turn modify the environment. By experimentally evaluating predictions made using these models we will test the degree to which our quantitative multiscale understanding wilt help to rationally steer individual microbes and their communities towards specific tasks. Towards this end we have made substantial progress towards understanding evolution of gene families, transcriptional structures, detailed structures of keystone molecular assemblies (proteins and complexes), protein interactions, biological networks, microbial interactions, and community structure. Using comparative analysis we have tracked the evolutionary history of gene functions to understand how novel functions evolve. One level up, we have used proteomics data, high-resolution genome tiling microarrays, and 5' RNA sequencing to revise genome annotations, discover new genes including ncRNAs, and map dynamically changing operon structures of five model organisms: For Desulfovibrio vulgaris Hildenborough, Pyrococcus furiosis, Sulfolobus solfataricus, Methanococcus maripaludis and Haiobacterium salinarum NROL We have developed machine learning algorithms to accurately identify protein interactions at a near-zero false positive rate from noisy data generated using tagfess complex purification, TAP purification, and analysis of membrane complexes. Combining other genome-scale datasets produced by ENIGMA (in particular, microarray data) and available from literature we have been able to achieve a true positive rate as high as 65% at almost zero false positives when applied to the manually curated training set. Applying this method to the data representing around a quarter of the fraction space for water soluble proteins in D. vulgaris, we obtained 854 reliable pair wise interactions. Further, we have developed algorithms to analyze and assign significance to protein interaction data from bait pull-down experiments and integrate these data with other systems biology data through associative biclustering in a parallel computing environment. We will 'fill-in' missing information in these interaction data using a 'Transitive Closure' algorithm and subsequently use 'Between Commonality Decomposition' algorithm to discover complexes within these large graphs of protein interactions. To characterize the metabolic activities of proteins and their complexes we are developing algorithms to deconvolute pure mass spectra, estimate chemical formula for m/z values, and fit isotopic fine structure to metabolomics data. We have discovered that in comparison to isotopic pattern fitting methods restricting the chemical formula by these two dimensions actually facilitates unique solutions for chemical formula generators. To understand how microbial functions are regulated we have developed complementary algorithms for reconstructing gene regulatory networks (GRNs). Whereas the network inference algorithms cMonkey and Inferelator developed enable de novo reconstruction of predictive models for GRNs from diverse systems biology data, the RegPrecise and RegPredict framework developed uses evolutionary comparisons of genomes from closely related organisms to reconstruct conserved regulons. We have integrated the two complementary algorithms to rapidly generate comprehensive models for gene regulation of understudied organisms. Our preliminary analyses of these reconstructed GRNs have revealed novel regulatory mechanisms and cis-regulatory motifs, as well asothers that are conserved across species. Finally, we are supporting scientific efforts in ENIGMA with data management solutions and by integrating all of the algorithms, software and data into a Knowledgebase. For instance, we have developed the RegPrecise database (http://regprecise.lbl.gov) which represents manually curated sets of regulons laying the basis for automatic annotation of regulatory interactions in closely related species. We are also in the midst of scaling up MicrobesOnline to handle the growing volume of sequence and functional genomics data. Over the last year our efforts have been focused on providing support for additional genomic and functional genomic data types. Similarly, we have developed several visualization tools to help with the exploration of complex systems biology datasets. A case in point is the Gaggle Genome Browser (GGB), which was enhanced with visualizations for plotting peptide detections and protein-DNA binding alongside transcriptome structure, plus the ability to interactively filter by signal intensity or p-value.

54 ENVIRONMENTAL SCIENCES↗

FLASH: FPGA-Accelerated Smart Switches with GCN Case Study

Some communication switches, e.g., the Mellanox SHArP and those in the IBM BlueGene clusters, are augmented to process packets at the application level with fixed-function collectives. This approach, however, lacks flexibility, which limits their applicability in diverse and dynamic workloads. Recently, a new type of programmable packet processor, which uses high-level languages, e.g., P4, has emerged as possible candidates. P4-based switches, however, fall short in certain applications, including machine learning, where capabilities not currently supported by P4 are needed. These include more complex calculation, such as sparse computation and fused multiply-accumulate, data-intensive floating point operations, data reuse, and significant memory. The problem addressed here is that such a switch augmentation needs to support: a large amount of state, significant flexible compute capability, and ease of programming, all while maintaining full functionality, including ensuring high throughput, and demonstrating utility. In this work, we propose a programmable look-aside-type accelerator that can be embedded into, or attached to, existing communication switch pipelines and that is capable of processing packets at line-rate. The proposed in-switch accelerator is based on mixing an ISA (subset of RISC-V instructions) with dataflow graphs (found in CGRAs). To augment performance, vector instructions are also supported. To facilitate usability, we have developed a complete toolchain to compile user-provided C/C++ codes to appropriate back-end instructions for configuring the accelerator. While this approach is flexible enough to support various workloads, in this paper, we consider Graph Convolutional Networks (GCNs) as a case study. Experimental results show that this approach considerably improves the performance of distributed GCN applications.

Haghi, Pouya↗

High-throughput predictions of metal–organic framework electronic properties: theoretical challenges, graph neural networks, and data exploration

Abstract With the goal of accelerating the design and discovery of metal–organic frameworks (MOFs) for electronic, optoelectronic, and energy storage applications, we present a dataset of predicted electronic structure properties for thousands of MOFs carried out using multiple density functional approximations. Compared to more accurate hybrid functionals, we find that the widely used PBE generalized gradient approximation (GGA) functional severely underpredicts MOF band gaps in a largely systematic manner for semi-conductors and insulators without magnetic character. However, an even larger and less predictable disparity in the band gap prediction is present for MOFs with open-shell 3 d transition metal cations. With regards to partial atomic charges, we find that different density functional approximations predict similar charges overall, although hybrid functionals tend to shift electron density away from the metal centers and onto the ligand environments compared to the GGA point of reference. Much more significant differences in partial atomic charges are observed when comparing different charge partitioning schemes. We conclude by using the dataset of computed MOF properties to train machine-learning models that can rapidly predict MOF band gaps for all four density functional approximations considered in this work, paving the way for future high-throughput screening studies. To encourage exploration and reuse of the theoretical calculations presented in this work, the curated data is made publicly available via an interactive and user-friendly web application on the Materials Project.

36 MATERIALS SCIENCE↗

Application-level benchmarking of quantum computers using nonlocal game strategies

In a nonlocal game, two noncommunicating players cooperate to convince a referee that they possess a strategy that does not violate the rules of the game. Quantum strategies allow players to optimally win some games by performing joint measurements on a shared entangled state, but computing these strategies can be challenging. We present a variational quantum algorithm to compute quantum strategies for nonlocal games by encoding the rules of a nonlocal game into a Hamiltonian. We show how this algorithm can generate a short-depth optimal quantum strategy for a graph coloring game with a quantum advantage. This quantum strategy is then evaluated on fourteen different quantum hardware platforms to demonstrate its utility as a benchmark. Finally, we discuss potential sources of errors that can explain the observed decreased performance of the executed task and derive an expression for the number of samples required to accurately estimate the win rate in the presence of noise.

nonlocal games↗

Randomized Algorithms for Symmetric Nonnegative Matrix Factorization

Symmetric Nonnegative Matrix Factorization (SymNMF) is a technique in data analysis and machine learning that approximates a matrix with a product of a nonnegative, low-rank matrix and it transpose. To design faster and more scalable algorithms for SymNMF we develop two randomized algorithms for its computation. The first method uses randomized matrix sketching to compute an initial low-rank approximation to the input matrix and proceeds to uses this as a low-rank input to rapidly compute a SymNMF. The second methods uses randomized leverage score sampling to approximately solve constrained least squares problems. Many successful methods for SymNMF rely on (approximately) solving sequences of constrained least squares problems. Here, we prove theoretically that leverage score sampling can approximately solve constrained least squares problems to e-accuracy. Finally we demonstrate both methods work in practice by applying them to graph clustering tasks on large real world data sets. These experiments show that our methods approximately maintain solution quality and achieve significant speed ups for both large dense and large sparse problems.

97 MATHEMATICS AND COMPUTING↗

Enabling Scalable and Extensible Memory-mapped Datastores in Userspace

Exascale workloads are expected to incorporate data-intensive processing in close coordination with traditional physics simulations. These emerging scientific, data-analytics and machine learning applications need to access a wide variety of datastores in flat files and structured databases. Programmer productivity is greatly enhanced by mapping datastores into the application process's virtual memory space to provide a unified “in-memory” interface. Currently, memory mapping is provided by system software primarily designed for generality and reliability. However, scalability at high concurrency is a formidable challenge on exascale systems. Also, there is a need for extensibility to support new datastores potentially requiring HPC data transfer services. In this article, we present UMap , a scalable and extensible userspace service for memory-mapping datastores. Furthermore, through decoupled queue management, concurrency aware adaptation, and dynamic load balancing, UMap enables application performance to scale even at high concurrency. We evaluate UMap in data-intensive applications, including sorting, graph traversal, database operations, and metagenomic analytics. Our results show that UMap as a userspace service outperforms an optimized kernel-based service across a wide range of intra-node concurrency by 1.22-1.9 × . We performed two case studies to demonstrate UMap 's extensibility. First, a new datastore residing in remote memory is incorporated into UMap as an application-specific plugin. Second, we present a persistent memory allocator Metall built atop UMap for unified storage/memory.

97 MATHEMATICS AND COMPUTING↗

Automated Recognition of 3D Features in GPIR Images

A method of automated recognition of three-dimensional (3D) features in images generated by ground-penetrating imaging radar (GPIR) is undergoing development. GPIR 3D images can be analyzed to detect and identify such subsurface features as pipes and other utility conduits. Until now, much of the analysis of GPIR images has been performed manually by expert operators who must visually identify and track each feature. The present method is intended to satisfy a need for more efficient and accurate analysis by means of algorithms that can automatically identify and track subsurface features, with minimal supervision by human operators. In this method, data from multiple sources (for example, data on different features extracted by different algorithms) are fused together for identifying subsurface objects. The algorithms of this method can be classified in several different ways. In one classification, the algorithms fall into three classes: (1) image-processing algorithms, (2) feature- extraction algorithms, and (3) a multiaxis data-fusion/pattern-recognition algorithm that includes a combination of machine-learning, pattern-recognition, and object-linking algorithms. The image-processing class includes preprocessing algorithms for reducing noise and enhancing target features for pattern recognition. The feature-extraction algorithms operate on preprocessed data to extract such specific features in images as two-dimensional (2D) slices of a pipe. Then the multiaxis data-fusion/ pattern-recognition algorithm identifies, classifies, and reconstructs 3D objects from the extracted features. In this process, multiple 2D features extracted by use of different algorithms and representing views along different directions are used to identify and reconstruct 3D objects. In object linking, which is an essential part of this process, features identified in successive 2D slices and located within a threshold radius of identical features in adjacent slices are linked in a directed-graph data structure. Relative to past approaches, this multiaxis approach offers the advantages of more reliable detections, better discrimination of objects, and provision of redundant information, which can be helpful in filling gaps in feature recognition by one of the component algorithms. The image-processing class also includes postprocessing algorithms that enhance identified features to prepare them for further scrutiny by human analysts (see figure). Enhancement of images as a postprocessing step is a significant departure from traditional practice, in which enhancement of images is a preprocessing step.

Park, Han↗

Parsl: Pervasive Parallel Programming in Python

High-level programming languages such as Python are increasingly used to provide intuitive interfaces to libraries written in lower-level languages and for assembling applications from various components. This migration towards orchestration rather than implementation, coupled with the growing need for parallel computing (e.g., due to big data and the end of Moore's law), necessitates rethinking how parallelism is expressed in programs. Here, we present Parsl, a parallel scripting library that augments Python with simple, scalable, and flexible constructs for encoding parallelism. These constructs allow Parsl to construct a dynamic dependency graph of components that it can then execute efficiently on one or many processors. Parsl is designed for scalability, with an extensible set of executors tailored to different use cases, such as low-latency, high-throughput, or extreme-scale execution. We show, via experiments on the Blue Waters supercomputer, that Parsl executors can allow Python scripts to execute components with as little as 5 ms of overhead, scale to more than 250000 workers across more than 8000 nodes, and process upward of 1200 tasks per second. Other Parsl features simplify the construction and execution of composite programs by supporting elastic provisioning and scaling of infrastructure, fault-tolerant execution, and integrated wide-area data management. We show that these capabilities satisfy the needs of many-task, interactive, online, and machine learning applications in fields such as biology, cosmology, and materials science.

parallel programming↗