Search NASA⌕ Search

SEARCH · Search NASA

Results for “computer system benchmarking”

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 613 records · Page 34

Designing Topology-Aware Collective Communication Algorithms for Large Scale InfiniBand Clusters: Case Studies with Scatter and Gather

Modern high performance computing systems are being increasingly deployed in a hierarchical fashion with multi-core computing platforms forming the base of the hierarchy. These systems are usually comprised of multiple racks, with each rack consisting of a finite number of chassis, with each chassis having multiple compute nodes or blades, based on multi-core architectures. The networks are also hierarchical with multiple levels of switches. Message exchange operations between processes that belong to different racks involve multiple hops across different switches and this directly affects the performance of collective operations. In this paper, we take on the challenges involved in detecting the topology of large scale InfiniBand clusters and leveraging this knowledge to design efficient topology-aware algorithms for collective operations. We also propose a communication model to analyze the communication costs involved in collective operations on large scale supercomputing systems. We have analyzed the performance characteristics of two collectives, MPI_Gather and MPI_Scatter on such systems and we have proposed topology-aware algorithms for these operations. Our experimental results have shown that the proposed algorithms can improve the performance of these collective operations by almost 54% at the micro-benchmark level.

97 MATHEMATICS AND COMPUTING↗

IOrchestrator: improving the performance of multi-node I/O systems via inter-server coordination

A cluster of I/O nodes and a parallel file system are often used to provide high-throughput I/O service to a parallel compute cluster. To exploit I/O parallelism parallel file systems stripe file data across the I/O nodes. While this practice is effective in serving asynchronous requests, it may break individual program's spatial locality, which can seriously degrade I/O performance when the I/O nodes concurrently serve synchronous requests from multiple I/O-intensive programs. In this paper we propose a scheme, Orchestrator, to improve the I/O performance of multi-I/O-node systems by orchestrating I/O services among programs when such inter-I/O-node coordination is dynamically determined to be cost effective. We have implemented IOrchestrator in the PVFS2 parallel file system. Our experiments with representative parallel benchmarks show that IOrchestrator can significantly improve I/O performance - by up to a factor of 2.5 - delivered by a cluster of I/O nodes servicing concurrently-running parallel programs. Notably, we have not observed any scenarios in which the use of IOrchestrator causes significant performance degradation.

97 MATHEMATICS AND COMPUTING↗

Center for Programming Models for Scalable Parallel Computing (Final Report)

As part of the Center for Programming Models for Scalable Parallel Computing, Rice University collaborated with project partners in the design, development and deployment of language, compiler, and runtime support for parallel programming models to support application development for the “leadership-class” computer systems at DOE national laboratories. Work over the course of this project has focused on the design, implementation, and evaluation of a second-generation version of Coarray Fortran. Research and development efforts of the project have focused on the CAF 2.0 language, compiler, runtime system, and supporting infrastructure. This has involved working with the teams that provide infrastructure for CAF that we rely on, implementing new language and runtime features, producing an open source compiler that enabled us to evaluate our ideas, and evaluating our design and implementation through the use of benchmarks. The report details the research, development, findings, and conclusions from this work.

97 MATHEMATICS AND COMPUTING↗

Keeping LAMMPS cutting edge

Since its inception 30 years ago, LAMMPS has grown to be a world-class molecular dynamics code and a cornerstone of computational materials science research. This project aimed to keep LAMMPS at the forefront of molecular dynamics simulations by adapting LAMMPS to the latest developments in machine learning technology and hardware. Initially, the project set out to provide a unified implementation of active learning for efficient training data generation in LAMMPS, but the research trajectory pivoted to address more immediate and impactful opportunities. On the hardware side, recent record-breaking molecular dynamics simulations were developed on the Cerebras wafer-scale AI chip, and this project has developed an interface between LAMMPS and the hardware-specific molecular dynamics code to accelerate and simplify development and user adoption. On the software side, PyTorch’s Ahead-of-Time (AOT) compilation features promised increased performance for state-of-the-art equivariant neural network potentials, and this project laid the groundwork for their adoption in LAMMPS, resulting in a nearly 20x acceleration in extreme cases. Combined with a comprehensive benchmark study of LAMMPS across all current exascale systems, this project has reinforced LAMMPS’s role as a versatile, high-performance tool for current and future materials science applications.

36 MATERIALS SCIENCE↗

Hierarchical fractional-step approximations and parallel kinetic Monte Carlo algorithms

We present a mathematical framework for constructing and analyzing parallel algorithms for lattice kinetic Monte Carlo (KMC) simulations. The resulting algorithms have the capacity to simulate a wide range of spatio-temporal scales in spatially distributed, non-equilibrium physiochemical processes with complex chemistry and transport micro-mechanisms. Rather than focusing on constructing exactly the stochastic trajectories, our approach relies on approximating the evolution of observables, such as density, coverage, correlations and so on. More specifically, we develop a spatial domain decomposition of the Markov operator (generator) that describes the evolution of all observables according to the kinetic Monte Carlo algorithm. This domain decomposition corresponds to a decomposition of the Markov generator into a hierarchy of operators and can be tailored to specific hierarchical parallel architectures such as multi-core processors or clusters of Graphical Processing Units (GPUs). Based on this operator decomposition, we formulate parallel Fractional step kinetic Monte Carlo algorithms by employing the Trotter Theorem and its randomized variants; these schemes, (a) are partially asynchronous on each fractional step time-window, and (b) are characterized by their communication schedule between processors. The proposed mathematical framework allows us to rigorously justify the numerical and statistical consistency of the proposed algorithms, showing the convergence of our approximating schemes to the original serial KMC. The approach also provides a systematic evaluation of different processor communicating schedules. We carry out a detailed benchmarking of the parallel KMC schemes using available exact solutions, for example, in Ising-type systems and we demonstrate the capabilities of the method to simulate complex spatially distributed reactions at very large scales on GPUs. Finally, we discuss work load balancing between processors and propose a re-balancing scheme based on probabilistic mass transport methods.

97 MATHEMATICS AND COMPUTING↗

Computational Challenges of Large-Scale Long-Time First-Principles Molecular Dynamics

Plane-wave density functional calculations have traditionally been able to use the largest available supercomputing resources. We analyze the scalability of modern projector- augmented wave implementations to identify the challenges in performing molecular dynamics calculations of large systems containing many thousands of electrons. Benchmark calculations on the Cray XT4 demonstrate that global linear-algebra operations are the primary reason for limited parallel scalability. Plane-wave related operations can be made sufficiently scalable. Improving parallel linear-algebra performance is an essential step to reaching longer timescales in future large-scale molecular dynamics calculations.

Kent, Paul↗

Strategic Plan for Nuclear Energy – Knowledge Base for Advanced Modeling and Simulation (NE-KAMS)

NE-KAMS knowledge base will assist computational analysts, physics model developers, experimentalists, nuclear reactor designers, and federal regulators by: (1) Establishing accepted standards, requirements and best practices for V&V and UQ of computational models and simulations, (2) Establishing accepted standards and procedures for qualifying and classifying experimental and numerical benchmark data, (3) Providing readily accessible databases for nuclear energy related experimental and numerical benchmark data that can be used in V&V assessments and computational methods development, (4) Providing a searchable knowledge base of information, documents and data on V&V and UQ, and (5) Providing web-enabled applications, tools and utilities for V&V and UQ activities, data assessment and processing, and information and data searches. From its inception, NE-KAMS will directly support nuclear energy research, development and demonstration programs within the U.S. Department of Energy (DOE), including the Consortium for Advanced Simulation of Light Water Reactors (CASL), the Nuclear Energy Advanced Modeling and Simulation (NEAMS), the Light Water Reactor Sustainability (LWRS), the Small Modular Reactors (SMR), and the Next Generation Nuclear Power Plant (NGNP) programs. These programs all involve computational modeling and simulation (M&S) of nuclear reactor systems, components and processes, and it is envisioned that NE-KAMS will help to coordinate and facilitate collaboration and sharing of resources and expertise for V&V and UQ across these programs. In addition, from the outset, NE-KAMS will support the use of computational M&S in the nuclear industry by developing guidelines and recommended practices aimed at quantifying the uncertainty and assessing the applicability of existing analysis models and methods. The NE-KAMS effort will initially focus on supporting the use of computational fluid dynamics (CFD) and thermal hydraulics (T/H) analysis for M&S of nuclear reactor systems, components and processes, and will later expand to include materials, fuel system performance and other areas of M&S as time and funding allow.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

An Ensemble Approach to Computationally Efficient Radiological Anomaly Detection and Isotope Identification

Radiological source search is a challenging task involving detection and identification of weak sources in a constantly changing radiological background. As of now, many radiological source detection algorithms have been proposed; however, their computational complexity, and hence reliance on power intensive processing units inhibit low-power applications of radiological source search systems. In this work, we introduce the anomaly filter (AF) algorithm; a computationally light, yet effective time-series source detection algorithm based on exponential weighted moving average (EWMA) and Poisson deviance statistics. Then, we demonstrate that the proposed algorithm can be used in ensemble with other more computationally intensive source detection and identification algorithms to achieve both increased detection performance and reduced power consumption. The proposed AF algorithm and the ensemble algorithms were thoroughly benchmarked against several existing source detection and identification algorithms. The results show that the AF algorithm outperforms existing conventional source detection algorithms, and the ensemble approach improves the overall performance of existing source detection and isotope identification algorithms. Furthermore, the AF algorithm and the Non-negative Matrix Factorization approach based source identification (NMF-ID) algorithm were combined and implemented on a singleboard microcontroller and the power consumption was measured. This ensemble algorithm reduced the power consumption of the NMF-ID algorithm almost by a factor of 100, while improving the detection performance of the overall system.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

CasCollect: targeted assembly of CRISPR-associated operons from high-throughput sequencing data

Abstract CRISPR arrays and CRISPR-associated (Cas) proteins comprise a widespread adaptive immune system in bacteria and archaea. These systems function as a defense against exogenous parasitic mobile genetic elements that include bacteriophages, plasmids and foreign nucleic acids. With the continuous spread of antibiotic resistance, knowledge of pathogen susceptibility to bacteriophage therapy is becoming more critical. Additionally, gene-editing applications would benefit from the discovery of new cas genes with favorable properties. While next-generation sequencing has produced staggering quantities of data, transitioning from raw sequencing reads to the identification of CRISPR/Cas systems has remained challenging. This is especially true for metagenomic data, which has the highest potential for identifying novel cas genes. We report a comprehensive computational pipeline, CasCollect, for the targeted assembly and annotation of cas genes and CRISPR arrays—even isolated arrays—from raw sequencing reads. Benchmarking our targeted assembly pipeline demonstrates significantly improved timing by almost two orders of magnitude compared with conventional assembly and annotation, while retaining the ability to detect CRISPR arrays and cas genes. CasCollect is a highly versatile pipeline and can be used for targeted assembly of any specialty gene set, reconfigurable for user provided Hidden Markov Models and/or reference nucleotide sequences.

Podlevsky, Joshua D.↗

: A Scalable and Transparent System for Simulating MPI Programs

is a scalable, transparent system for experimenting with the execution of parallel programs on simulated computing platforms. The level of simulated detail can be varied for application behavior as well as for machine characteristics. Unique features of are repeatability of execution, scalability to millions of simulated (virtual) MPI ranks, scalability to hundreds of thousands of host (real) MPI ranks, portability of the system to a variety of host supercomputing platforms, and the ability to experiment with scientific applications whose source-code is available. The set of source-code interfaces supported by is being expanded to support a wider set of applications, and MPI-based scientific computing benchmarks are being ported. In proof-of-concept experiments, has been successfully exercised to spawn and sustain very large-scale executions of an MPI test program given in source code form. Low slowdowns are observed, due to its use of purely discrete event style of execution, and due to the scalability and efficiency of the underlying parallel discrete event simulation engine, sik. In the largest runs, has been executed on up to 216,000 cores of a Cray XT5 supercomputer, successfully simulating over 27 million virtual MPI ranks, each virtual rank containing its own thread context, and all ranks fully synchronized by virtual time.

97 MATHEMATICS AND COMPUTING↗

An efficient MPI/OpenMP parallelization of the Hartree–Fock–Roothaan method for the first generation of Intel® Xeon Phi™ processor architecture

The Hartree-Fock (HF) method in the quantum chemistry package GAMESS represents one of the most irregular algorithms in computation today. Major steps in the calculation are the irregular computation of electron repulsion integrals (ERIs) and the building of the Fock matrix. These are the central components of the main Self Consistent Field (SCF) loop, the key hotspot in Electronic Structure (ES) codes. By threading the MPI ranks in the official release of the GAMESS code, we not only speed up the main SCF loop (4x to 6x for large systems), but also achieve a significant (>2x) reduction in the overall memory footprint. These improvements are a direct consequence of memory access optimizations within the MPI ranks. We benchmark our implementation against the official release of the GAMESS code on the Intel R Xeon PhiTM supercomputer. Here, scaling numbers are reported on up to 7,680 cores on Intel Xeon Phi coprocessors.

97 MATHEMATICS AND COMPUTING↗

Workshop on Advanced Computing for Connected and Automated Vehicles

To safely and reliably operate without a human driver, connected and automated vehicles (CAVs) require more advanced computing hardware and software solutions than are implemented today in vehicles that provide driver-assistance features. A workshop was held to discuss advanced microelectronics and computing approaches that can help meet future energy and computational requirements for CAVs. Workshop questions were posed as follows: will highly automated vehicles be viable with conventional computing approaches or will they require a step-change in computing; what are the energy requirements to support on-board sensing and computing; and what advanced computing approaches could reduce the energy requirements while meeting their computational requirements? At present, there is no clear convergence in the computing architecture for highly automated vehicles. However, workshop participants generally agreed that there is a need to improve the computing performance per watt by at least 10x to advance the degree of automation. Participants suggested that DOE and the national laboratories could play a near-term role by developing benchmarks for determining and comparing CAV computing performance, developing public data sets to support algorithm and software development, and contributing precompetitive advancements in energy efficient computing.

33 ADVANCED PROPULSION SYSTEMS↗

Development and testing of neutron cross-section covariance data for Scale 6.2 - 14604

Neutron cross-section covariance data are essential for many sensitivity/uncertainty and uncertainty quantification assessments performed both within the TSUNAMI suite and more broadly throughout the SCALE code system. The release of ENDF/B-VII.1 included a more complete set of neutron cross-section covariance data: these data form the basis for a new cross-section covariance library to be released in SCALE 6.2. A range of testing is conducted to investigate the properties of these covariance data and ensure that the data are reasonable. These tests include examination of the uncertainty in critical experiment benchmark model k{sub eff} values due to nuclear data uncertainties, as well as similarity assessments of irradiated pressurized water reactor (PWR) and boiling water reactor (BWR) fuel with suites of critical experiments. The contents of the new covariance library, the testing performed, and the behavior of the new covariance data are described in this paper. The neutron cross-section covariances can be combined with a sensitivity data file generated using the TSUNAMI suite of codes within SCALE to determine the uncertainty in system k{sub eff} caused by nuclear data uncertainties. The Verified, Archived Library of Inputs and Data (VALID) maintained at Oak Ridge National Laboratory (ORNL) contains over 400 critical experiment benchmark models, and sensitivity data are generated for each of these models. The nuclear data uncertainty in k{sub eff} is generated for each experiment, and the resulting uncertainties are tabulated and compared to the differences in measured and calculated results. The magnitude of the uncertainty for categories of nuclides (such as actinides, fission products, and structural materials) is calculated for irradiated PWR and BWR fuel to quantify the effect of covariance library changes between the SCALE 6.1 and 6.2 libraries. One of the primary applications of sensitivity/uncertainty methods within SCALE is the assessment of similarities between benchmark experiments and safety applications. This is described by a ck value for each experiment with each application. Several studies have analyzed typical ck values for a range of critical experiments compared with hypothetical irradiated fuel applications. The ck value is sensitive to the cross-section covariance data because the contribution of each nuclide is influenced by its uncertainty; large uncertainties indicate more likely bias sources and are thus given more weight. Changes in ck values resulting from different covariance data can be used to examine and assess underlying data changes. These comparisons are performed for PWR and BWR fuel in storage and transportation systems. (authors)

42 ENGINEERING↗

A nonvoxel-based dose convolution/superposition algorithm optimized for scalable GPU architectures

Purpose: Real-time adaptive planning and treatment has been infeasible due in part to its high computational complexity. There have been many recent efforts to utilize graphics processing units (GPUs) to accelerate the computational performance and dose accuracy in radiation therapy. Data structure and memory access patterns are the key GPU factors that determine the computational performance and accuracy. In this paper, the authors present a nonvoxel-based (NVB) approach to maximize computational and memory access efficiency and throughput on the GPU. Methods: The proposed algorithm employs a ray-tracing mechanism to restructure the 3D data sets computed from the CT anatomy into a nonvoxel-based framework. In a process that takes only a few milliseconds of computing time, the algorithm restructured the data sets by ray-tracing through precalculated CT volumes to realign the coordinate system along the convolution direction, as defined by zenithal and azimuthal angles. During the ray-tracing step, the data were resampled according to radial sampling and parallel ray-spacing parameters making the algorithm independent of the original CT resolution. The nonvoxel-based algorithm presented in this paper also demonstrated a trade-off in computational performance and dose accuracy for different coordinate system configurations. In order to find the best balance between the computed speedup and the accuracy, the authors employed an exhaustive parameter search on all sampling parameters that defined the coordinate system configuration: zenithal, azimuthal, and radial sampling of the convolution algorithm, as well as the parallel ray spacing during ray tracing. The angular sampling parameters were varied between 4 and 48 discrete angles, while both radial sampling and parallel ray spacing were varied from 0.5 to 10 mm. The gamma distribution analysis method (γ) was used to compare the dose distributions using 2% and 2 mm dose difference and distance-to-agreement criteria, respectively. Accuracy was investigated using three distinct phantoms with varied geometries and heterogeneities and on a series of 14 segmented lung CT data sets. Performance gains were calculated using three 256 mm cube homogenous water phantoms, with isotropic voxel dimensions of 1, 2, and 4 mm. Results: The nonvoxel-based GPU algorithm was independent of the data size and provided significant computational gains over the CPU algorithm for large CT data sizes. The parameter search analysis also showed that the ray combination of 8 zenithal and 8 azimuthal angles along with 1 mm radial sampling and 2 mm parallel ray spacing maintained dose accuracy with greater than 99% of voxels passing the γ test. Combining the acceleration obtained from GPU parallelization with the sampling optimization, the authors achieved a total performance improvement factor of >175 000 when compared to our voxel-based ground truth CPU benchmark and a factor of 20 compared with a voxel-based GPU dose convolution method. Conclusions: The nonvoxel-based convolution method yielded substantial performance improvements over a generic GPU implementation, while maintaining accuracy as compared to a CPU computed ground truth dose distribution. Such an algorithm can be a key contribution toward developing tools for adaptive radiation therapy systems.

60 APPLIED LIFE SCIENCES↗

Genetic algorithm based task reordering to improve the performance of batch scheduled massively parallel scientific applications

The growth in size of networked high performance computers along with novel accelerator-based node architectures has further emphasized the importance of communication efficiency in high performance computing. The world's largest high performance computers are usually operated as shared user facilities due to the costs of acquisition and operation. Applications are scheduled for execution in a shared environment and are placed on nodes that are not necessarily contiguous on the interconnect. Furthermore, the placement of tasks on the nodes allocated by the scheduler is sub-optimal, leading to performance loss and variability. Here, we investigate the impact of task placement on the performance of two massively parallel application codes on the Titan supercomputer, a turbulent combustion flow solver (S3D) and a molecular dynamics code (LAMMPS). Benchmark studies show a significant deviation from ideal weak scaling and variability in performance. The inter-task communication distance was determined to be one of the significant contributors to the performance degradation and variability. A genetic algorithm-based parallel optimization technique was used to optimize the task ordering. This technique provides an improved placement of the tasks on the nodes, taking into account the application's communication topology and the system interconnect topology. As a result, application benchmarks after task reordering through genetic algorithm show a significant improvement in performance and reduction in variability, therefore enabling the applications to achieve better time to solution and scalability on Titan during production.

97 MATHEMATICS AND COMPUTING↗

Time-dependent projected Hartree-Fock

Projected Hartree-Fock (PHF) has recently emerged as an alternative approach to describing degenerate systems where static correlation is abundant, when the spin-symmetry is projected. Here, we derive a set of linearized time-dependent equations for PHF in order to be able to access excited states. The close connection of such linear-response time-dependent PHF (TDPHF) to the stability condition of a PHF wave function is discussed. Expanding this analysis also makes it possible to give analytical expressions for the projected coupling terms of Hamiltonian and overlaps between excited Slater determinants. TDPHF with spin-projection (TDSUHF) and its Tamm-Dancoff approximation are benchmarked for several electronically degenerate molecules including the dissociating H{sub 2}, F{sub 2} and O{sub 3} at equilibrium, and the distorted ethylene. It is shown that they give consistently better descriptions of excited states than does time-dependent HF (TDHF). Furthermore, we demonstrate that they offer not only singly but also doubly excited states, which naturally arise upon spin-projection. We also address the thermodynamic limit of TDSUHF, using non-interacting He gas. While TDPHF singly excited states tend to converge to those of HF with the size of the system due to the lack of size-extensivity of PHF, doubly excited states remain reasonable even at the thermodynamic limit. We find that the overall performance of our method is systematically better than the regular TDHF in many cases at the same computational scaling.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Collective input/output under memory constraints

Compared with current high-performance computing (HPC) systems, exascale systems are expected to have much less memory per node, which can significantly reduce necessary collective input/output (I/O) performance. In this study, we introduce a memory-conscious collective I/O strategy that takes into account memory capacity and bandwidth constraints. The new strategy restricts aggregation data traffic within disjointed subgroups, coordinates I/O accesses in intranode and internode layers, and determines I/O aggregators at run time considering memory consumption among processes. We have prototyped the design and evaluated it with commonly used benchmarks to verify its potential. The evaluation results demonstrate that this strategy holds promise in mitigating the memory pressure, alleviating the contention for memory bandwidth, and improving the I/O performance for projected extreme-scale systems. Given the importance of supporting increasingly data-intensive workloads and projected memory constraints on increasingly larger scale HPC systems, this new memory-conscious collective I/O can have a significant positive impact on scientific discovery productivity.

collective input/output↗

Chimbuko: A Workflow-Level Scalable Performance Trace Analysis Tool

Due to the sheer volume of data it is typically impractical to analyze the detailed performance of an HPC application running at-scale. While conventional small-scale benchmarking and scaling studies are often sufficient for simple applications, many modern workflow-based applications couple multiple elements with competing resource demands and complex inter-communication patterns for which performance cannot easily be studied in isolation and at small scale. This work discusses Chimbuko, a performance analysis framework that provides real-time, in situ anomaly detection. By focusing specifically on performance anomalies and their origin (aka provenance), data volumes are dramatically reduced without losing necessary details. To the best of our knowledge, Chimbuko is the first online, distributed, and scalable workflow-level performance trace analysis framework. We demonstrate the tool's usefulness on Oak Ridge National Laboratory's Summit system.

97 MATHEMATICS AND COMPUTING↗