SHMEM-ML: Leveraging OpenSHMEM and Apache Arrow for Scalable, Composable Machine Learning
Explore the source record for details and available documents.
Computer Science↗
SEARCH · Engineering Papers
Results for “Scalable computation”
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.
Towards Scalable Data Processing in Python with CLIPPy
Explore the source record for details and available documents.
Computer science↗
Enhancing Scalability for FUN3D Rotorcraft Simulations with Yoga: an Overset Grid Assembler
FUN3D, an unstructured grid Navier-Stokes CFD code is capable of overset grid simu- lations, but does not have an internal method for assembling an overset grid system from a group of component grids. FUN3D currently relies on the third party codes Suggar++ and DiRTlib to perform domain assembly and provide intergrid connectivity for overset simulations. Rotorcraft simulations with moving, deforming blades require domain assembly and mesh deformation at each time step. For these simulations, the three primary drivers of computational cost for each time step are: deforming the mesh, performing domain assembly, and performing subiterations of the flow solver. FUN3D exhibits strong and weak scalability for the flow solver subiterations and the mesh deformation. However, FUN3D is currently hardwired directly to the serial version of Suggar++, which has a fixed cost for a given mesh system. Therefore, domain assembly begins to dominate the total cost of each time step as grid systems become larger. An integrated method for parallel domain assembly is presented that addresses scalability for large grid systems. Restructuring within FUN3D to accommodate integrated domain assembly is also discussed, which could enable use of the parallel Suggar++ library.
Cameron T Druyor↗
A Scalable Multi-Modal Framework for High-Fidelity Distributed Human Mobility Simulations
The development of data-driven models for human mobility in urban settings requires access to substantial and diverse real-world data. However, existing historical data often presents challenges such as limited volume, variety, and veracity, as well as missing data and privacy preservation concerns. Also, urban mobility modeling is inherently time-variant, complex, and multi-modal, encompassing everything from individual walking and running to private road travel and large-scale public transportation. These challenges call for innovative solutions to overcome data limitations and compute needs to model mobility behaviors accurately. To address these challenges, we propose a distributed, co-simulation-based architecture DURMOSim that integrates real-world data with scalable, high-fidelity simulations, demonstrating distributed co-simulation feasibility with existing mobility models. DURMOSim underpins a modular integration that would enable using any available mobility simulators for greater extensibility and scalability in performing various urban scenarios. In this paper, we present the design, implementation, and performance evaluation of DURMOSim, highlighting its capability to model population-scale mobility patterns. Our initial results show its ability to dynamically synchronize multiple simulation models at runtime with negligible computational overhead. We believe DURMOSim could be a robust tool for advancing urban mobility research and intelligent transportation systems.
Yoginath, Srikanth [ORNL] (ORCID:0000000184236050)↗
Transition metal impurities in silicon: computational search for a semiconductor qubit
Abstract Semiconductors offer a promising platform for physical implementation of qubits, but their broad adoption is presently hindered by limited scalability and/or very low operating temperatures. Learning from the nitrogen-vacancy centers in diamond, our goal is to find equivalent optically active point defect centers in crystalline silicon, which could be advantageous for their scalability and integration with classical devices. Transition metal (TM) impurities in silicon are common paramagnetic deep defects, but a comprehensive theoretical study of the whole 3 d series that considers generalized Koopmans’ condition is missing. We apply the HSE06(+U) method to examine their potential as optically active spin qubits and identify seven TM impurities that have optically allowed triplet–triplet transitions within the silicon band gap. These results provide the first step toward silicon-based qubits with higher operating temperatures for quantum sensing. Additionally, these point defects could lead to spin-photon interfaces in silicon-based qubits and devices for mid-infrared free-space communications.
36 MATERIALS SCIENCE↗
Enabling Exascale Hardware and Software Design through Scalable System Virtualization (Final Report)
The purpose of this project has been to extend the state of the art of systems software for high-end computing (HEC) platforms, and to use systems software to better enable the evaluation of potential future HEC platforms, for example exascale platforms. Such platforms, and their systems software, have the goal of providing scientific computation at new scales, thus enabling new research in the physical sciences and engineering. Over time, the innovations in systems software for such platforms also become applicable to more widely used computing clusters, data centers, and clouds. This was a five-institution project, centered on the Palacios virtual machine monitor (VMM) systems software, a project begun at Northwestern, and originally developed in a previous collaboration between Northwestern University and the University of New Mexico. In this project, Northwestern (including via our subcontract to the University of Pittsburgh) contributed to the continued development of Palacios, along with other team members. We took the leadership role in (1) continued extension of support for emerging Intel and AMD hardware, (2) integration and performance enhancement of overlay networking, (3) connectivity with architectural simulation, (4) binary translation, and (5) support for modern Non-Uniform Memory Access (NUMA) hosts and guests. We also took a supporting role in support for specialized hardware for I/O virtualization, profiling, configurability, and integration with configuration tools. The efforts we led (1-5) were largely successful and executed as expected, with code and papers resulting from them. The project demonstrated the feasibility of a virtualization layer for HEC computing, similar to such layers for cloud or datacenter computing. For effort (3), although a prototype connecting Palacios with the GEM5 architectural simulator was demonstrated, our conclusion was that such a platform was less useful for design space exploration than anticipated due to inherent complexity of the connection between the instruction set architecture level and the microarchitectural level. For effort (4), we found that a code injection approach proved to be more fruitful. The results of our efforts are publicly available in the open source Palacios codebase and published papers, all of which are available from the project web site, v3vee.org. Palacios is currently one of the two codebases (the other being Sandia’s Kitten lightweight kernel) that underlies the node operating system for the DOE Hobbes Project, one of two projects tasked with building a systems software prototype for the national exascale computing effort.
97 MATHEMATICS AND COMPUTING↗
Synthetic Graph Generation for Data-Intensive HPC Benchmarking: Scalability, Analysis and Real-World Application
The benchmarking effort within the Extreme Scale Systems Center at Oak Ridge National Laboratory seeks to provide High Performance Computing benchmarks and test suites of interest to the DoD sponsor. The work described in this report is a part of the effort focusing on graph generation. A previously developed benchmark, SystemBurn, allows the emulation of a broad spectrum of application behavior profiles within a single framework. To complement this effort, similar capabilities are desired for graph-centric problems. This report described the in-depth analysis of the generated synthetic graphs' properties at a variety of scales using different generator implementations and examines their applicability to replicating real world datasets.
97 MATHEMATICS AND COMPUTING↗
Final Report Scalable Analysis Methods and In Situ Infrastructure for Extreme Scale Knowledge Discovery
The primary challenge motivating this project is the widening gap between the ability to compute information and to store it for subsequent analysis. This gap adversely impacts science code teams, who can perform analysis only on a small fraction of the data they calculate, resulting in the substantial likelihood of lost or missed science, when results are computed but not analyzed. Our approach is to perform as much analysis or visualization processing on data while it is still resident in memory, which is known as in situ processing. The idea in situ processing was not new at the time of the start of this effort in 2014, but efforts in that space were largely ad hoc, and there was no concerted effort within the research community that aimed to foster production-quality software tools suitable for use by Department of Energy (DOE) science projects. Our objective was to produce and enable the use of production-quality in situ methods and infrastructure, at scale, on DOE high-performance computing (HPC) facilities, though we expected to have an impact beyond DOE due to the widespread nature of the challenges, which affect virtually all large-scale computational science efforts. To achieve this objective, we engaged in software technology research and development (R&D), in close partnerships with DOE science code teams, to produce software technologies that were shown to run efficiently at scale on DOE HPC platforms.
96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
Scalable All-pairs Shortest Paths for Huge Graphs on Multi-GPU Clusters
We present an optimized Floyd-Warshall (Floyd-Warshall) algorithm that computes the All-pairs shortest path (APSP) for GPU accelerated clusters. The Floyd-Warshall algorithm due to its structural similarities to matrix-multiplication is well suited for highly parallel GPU architectures. To achieve high parallel efficiency, we address two key algorithmic challenges: reducing high communication overhead and addressing limited GPU memory. To reduce high communication costs, we redesign the parallel (a) to expose more parallelism, (b) aggressively overlap communication and computation with pipelined and asynchronous scheduling of operations, and (c) tailored MPI-collective. To cope with limited GPU memory, we employ an offload model, where the data resides on the host and is transferred to GPU on-demand. The proposed optimizations are supported with detailed performance models for tuning. Our optimized parallel Floyd-Warshall implementation is up to 5x faster than a strong baseline and achieves 8.1 PetaFLOPS/sec on 256~nodes of the Summit supercomputer at Oak Ridge National Laboratory. This performance represents 70% of the theoretical peak and 80% parallel efficiency. The offload algorithm can handle 2.5x larger graphs with a 20% increase in overall running time.
Sao, Piyush↗
Automatic Detection of Defects in High-Reliability Components
Disastrous consequences can result from defects in manufactured parts—particularly the high consequence parts developed at Sandia. Identifying flaws in as-built parts can be done with nondestructive means, such as X-ray Computed Tomography (CT). However, due to artifacts and complex imagery, the task of analyzing the CT images falls to humans. Human analysis is inherently unreproducible, unscalable, and can easily miss subtle flaws. We hypothesized that deep learning methods could improve defect identification, increase the number of parts that can effectively be analyzed, and do it in a reproducible manner. We pursued two methods: 1) generating a defect-free version of a scan and looking for differences (PandaNet), and 2) using pre-trained models to develop a statistical model of normality (Feature-based Anomaly Detection System: FADS). Both PandaNet and FADS provide good results, are scalable, and can identify anomalies in imagery. In particular, FADS enables zero-shot (training-free) identification of defects for minimal computational cost and expert time. It significantly outperforms prior approaches in computational cost while achieving comparable results. FADS’ core concept has also shown utility beyond anomaly detection by providing feature extraction for downstream tasks.
47 OTHER INSTRUMENTATION↗
Integrating PGAS and MPI-based Graph Analysis
This project demonstrates that Chapel programs can interface with MPI-based libraries written in C++ without storing multiple copies of shared data. Chapel is a language for productive parallel computing using global address spaces (PGAS). We identified two approaches to interface Chapel code with the MPI-based Grafiki and Trilinos libraries. The first uses a single Chapel executable to call a C function that interacts with the C++ libraries. The second uses the mmap function to allow separate executables to read and write to the same block of memory on a node. We also encapsulated the second approach in Docker/Singularity containers to maximize ease of use. Comparisons of the two approaches using shared and distributed memory installations of Chapel show that both approaches provide similar scalability and performance.
97 MATHEMATICS AND COMPUTING↗
The uintah framework: a unified heterogeneous task scheduling and runtime system
The development of a new unified, multi-threaded runtime system for the execution of asynchronous tasks on heterogeneous systems is described in this work. These asynchronous tasks arise from the Uintah framework, which was developed to provide an environment for solving a broad class of fluid-structure interaction problems on structured adaptive grids. Uintah has a clear separation between its MPI-free user-coded tasks and its runtime system that ensures these tasks execute efficiently. This separation also allows for complete isolation of the application developer from the complexities involved with the parallelism Uintah provides. While we have designed scalable runtime systems for large CPU core counts, the emergence of heterogeneous systems, with additional on-node accelerators and co-processors presents additional design challenges in terms of effectively utilizing all computational resources on-node and managing multiple levels of parallelism. Our work addresses these challenges for Uintah by the development of new hybrid runtime system and Unified multi-threaded MPI task scheduler, enabling Uintah to fully exploit current and emerging architectures with support for asynchronous, out-of-order scheduling of both CPU and GPU computational tasks. This design coupled with an approach that uses MPI to communicate between nodes, a shared memory model on-node and the use of novel lock-free data structures, has made it possible for Uintah to achieve excellent scalability for challenging fluid-structure problems using adaptive mesh refinement on as many as 256K cores on the DoE Jaguar XK6 system. This design has also demonstrated an ability to run capability jobs on the heterogeneous systems, Keeneland and TitanDev. In this work, the evolution of Uintah and its runtime system is examined in the context of our new Unified multi-threaded scheduler design. The performance of the Unified scheduler is also tested against previous Uintah scheduler and runtime designs over a range of processor core and GPU counts.
Source record↗
DeepHyper: Asynchronous Hyperparameter Search for Deep Neural Networks
Hyperparameters employed by deep learning (DL) methods play a substantial role in the performance and reliability of these methods in practice. Unfortunately, finding performance-optimizing hyperparameter settings is a notoriously difficult task. Hyperparameter search methods typically have limited production-strength implementations or do not target scalability within a highly parallel machine, portability across different machines, experimental comparison between different methods, and tighter integration with workflow systems. In this paper, we present DeepHyper, a Python package that provides a common interface for the implementation and study of scalable hyperparameter search methods. It adopts the Balsam workflow system to hide the complexities of running large numbers of hyperparameter configurations in parallel on high-performance computing (HPC) systems. We implement and study asynchronous model-based search methods that consist of sampling a small number of input hyperparameter configurations and progressively fitting surrogate models over the input-output space until exhausting a user-defined budget of evaluations. We evaluate the efficacy of these methods relative to approaches such as random search, genetic algorithms, Bayesian optimization, and hyperband on DL benchmarks on CPU- and GPU-based HPC systems.
Balaprakash, Prasanna↗
On the Efficient Evaluation of the Exchange Correlation Potential on Graphics Processing Unit Clusters
The predominance of Kohn–Sham density functional theory (KS-DFT) for the theoretical treatment of large experimentally relevant systems in molecular chemistry and materials science relies primarily on the existence of efficient software implementations which are capable of leveraging the latest advances in modern high-performance computing (HPC). With recent trends in HPC leading toward increasing reliance on heterogeneous accelerator-based architectures such as graphics processing units (GPU), existing code bases must embrace these architectural advances to maintain the high levels of performance that have come to be expected for these methods. In this work, we purpose a three-level parallelism scheme for the distributed numerical integration of the exchange-correlation (XC) potential in the Gaussian basis set discretization of the Kohn–Sham equations on large computing clusters consisting of multiple GPUs per compute node. In addition, we purpose and demonstrate the efficacy of the use of batched kernels, including batched level-3 BLAS operations, in achieving high levels of performance on the GPU. We demonstrate the performance and scalability of the implementation of the purposed method in the NWChemEx software package by comparing to the existing scalable CPU XC integration in NWChem.
97 MATHEMATICS AND COMPUTING↗
SUPG stabilization for the nonconforming virtual element method for advection–diffusion–reaction equations
We present here the design, convergence analysis and numerical investigations of the nonconforming virtual element method with Streamline Upwind/Petrov–Galerkin (VEM-SUPG) stabilization for the numerical resolution of convection–diffusion–reaction problems in the convective-dominated regime. According to the virtual discretization approach, the bilinear form is split as the sum of a consistency and a stability term. The consistency term is given by substituting the functions of the virtual space and their gradients with their polynomial projection in each term of the bilinear form (including the SUPG stabilization term). Polynomial projections can be computed exactly from the degrees of freedom. The stability term is also built from the degrees of freedom by ensuring the correct scalability properties with respect to the mesh size and the equation coefficients. The nonconforming formulation relaxes the continuity conditions at cell interfaces and a weaker regularity condition is considered involving polynomial moments of the solution jumps at cell interface. Optimal convergence properties of the method are proved in a suitable norm, which includes contribution from the advective stabilization terms. Experimental results confirm the theoretical convergence rates.
97 MATHEMATICS AND COMPUTING↗
PapyrusKV: a high-performance parallel key-value store for distributed NVM architectures
This paper introduces PapyrusKV, a parallel embedded key-value store (KVS) for distributed high-performance computing (HPC) architectures that offer potentially massive pools of nonvolatile memory (NVM). PapyrusKV stores keys with their values in arbitrary byte arrays across multiple NVMs in a distributed system. PapyrusKV provides standard KVS operations such as put, get, and delete. More importantly, PapyrusKV provides advanced features for HPC such as dynamic consistency control, zero-copy workflow, and asynchronous checkpoint/restart. Beyond filesystems, PapyrusKV provides HPC programmers with a high-level interface to exploit distributed NVM in the system, and it transparently organizes data to achieve high performance. Also, it allows HPC applications to specialize PapyrusKV to meet their specific requirements. We empirically evaluate PapyrusKV on three HPC systems with real NVM devices: OLCF's Summitdev, TACC's Stampede, and NERSC's Cori. Our results show that PapyrusKV can offer high performance, scalability, and portability across these various distributed NVM architectures.
97 MATHEMATICS AND COMPUTING↗
Enabling Reliable, Fault-Tolerant Autonomous Lunar Habitats with High-Performance Spaceflight Computing
The lunar surface presents unfavorable constraints and harsh living conditions. To address these challenges, autonomous habitats will require complex integrated systems that combine advanced software, high-performance hardware, and cutting-edge sensors to ensure sustainability, safety, and operational efficiency. Consequently, maintaining a sustainable presence on the Moon requires reliable infrastructure and efficient development, precise monitoring, and utilization of resources within a lunar installation. These elements are essential not only to ensure that lunar settlement can be long-term, self-sustaining, and resource-efficient, but also to serve as a foundation for future missions and eventual human habitation on Mars. Humans are not native to the Moon; therefore, our survival and ability to thrive will depend on autonomous systems that can foster safety and resilience through high-availability architectures, graceful degradation, and highly fault-tolerant spaceflight hardware capable of continuing operation during failures. This requires advanced human-rated distributed systems architectures with specialized electronics, scalable capabilities, and an integrated design approach. Unlike current practices focused on short-term missions and regularly maintained components, permanent lunar compute systems must be designed for extended operations beyond mission durations. This paper explores the necessity of transitioning toward fault- tolerant, highly autonomous hardware systems designed for multi-year missions. It also identifies critical subsystems that require high levels of autonomy, supported by radiation-hardened processors and extreme thermal loads, which are essential to mitigate long-term degradation and ensure sustainable lunar habitation. Finally, the paper aligns with NASA’s identified Civil Space Shortfalls, particularly in high-performance onboard computing, advanced data acquisition, extreme-environment avionics, radiation monitoring and countermeasures, and autonomous health management. It proposes NASA’s new High-Performance Spaceflight Computing (HPSC) processor as a turnkey solution, delivering 100 times the performance-per-watt of legacy rad-hard CPUs and enabling onboard AI, edge computing, and fault-tolerant features essential for sustained lunar autonomy and beyond.
Sarkis S Mikaelian↗
Π4U: A high performance computing framework for Bayesian uncertainty quantification of complex models
We present Π4U,{sup 1} an extensible framework, for non-intrusive Bayesian Uncertainty Quantification and Propagation (UQ+P) of complex and computationally demanding physical models, that can exploit massively parallel computer architectures. The framework incorporates Laplace asymptotic approximations as well as stochastic algorithms, along with distributed numerical differentiation and task-based parallelism for heterogeneous clusters. Sampling is based on the Transitional Markov Chain Monte Carlo (TMCMC) algorithm and its variants. The optimization tasks associated with the asymptotic approximations are treated via the Covariance Matrix Adaptation Evolution Strategy (CMA-ES). A modified subset simulation method is used for posterior reliability measurements of rare events. The framework accommodates scheduling of multiple physical model evaluations based on an adaptive load balancing library and shows excellent scalability. In addition to the software framework, we also provide guidelines as to the applicability and efficiency of Bayesian tools when applied to computationally demanding physical models. Theoretical and computational developments are demonstrated with applications drawn from molecular dynamics, structural dynamics and granular flow.
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗