Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Parallel Supercomputers”

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 289 records · Page 16

Provisioning a Multi-Tiered Data Staging Area for Extreme-Scale Machines

Massively parallel scientific applications, running on extreme-scale supercomputers, produce hundreds of terabytes of data per run, driving the need for storage solutions to improve their I/O performance. Traditional parallel file systems (PFS) in high performance computing (HPC) systems are unable to keep up with such high data rates, creating a storage wall. In this work, we present a novel multi-tiered storage architecture comprising hybrid node-local resources to construct a dynamic data staging area for extreme-scale machines. Such a staging ground serves as an impedance matching device between applications and the PFS. Our solution combines diverse resources (e.g., DRAM, SSD) in such a way as to approach the performance of the fastest component technology and the cost of the least expensive one. We have developed an automated provisioning algorithm that aids in meeting the checkpointing performance requirement of HPC applications, by using a least-cost storage configuration. We evaluate our approach using both an implementation on a large scale cluster and a simulation driven by six-years worth of Jaguar supercomputer job-logs, and show that our approach, by choosing an appropriate storage configuration, achieves 41.5% cost savings with only negligible impact on performance.

99 GENERAL AND MISCELLANEOUS↗

Applications of the massively parallel machine, the MasPar MP-1, to Earth sciences

The computational workload of upcoming NASA science missions, especially the ground data processing for the Earth Observing System, is projected to be quite large (in the 50 to 100 gigaFLOPS range) and corespondingly very expensive to perform using conventional supercomputer systems. High performance, general purpose massively parallel computer systems such as the MasPar MP-1 are being investigated by NASA as a more cost effective alternative. Massively parallel systems are targeted for accelerated development and maturation by NASA's upcoming five-year High Performance Computing and Communications Program. A summary of the broad range of applications currently running on the MP-1 at NASA/Goddard are presented in this paper along with descriptions of the parallel algorithmic techniques employed in five applications that have bearing on Earth sciences.

Fischer, James R.↗

Designing Energy Efficient Communication Runtime Systems for Data Centric Programming Models

The insatiable demand of high performance computing is being driven by the most computationally intensive applications such as computational chemistry, climate modeling, nuclear physics, etc. The last couple of decades have observed a tremendous rise in supercomputers with architectures ranging from traditional clusters to system-on-a-chip in order to achieve the petaflop computing barrier. However, with advent of petaflop-plus computing, we have ushered in an era where power efficient system software stack is imperative for execution on exascale systems and beyond. At the same time, computationally intensive applications are exploring programming models beyond traditional message passing, as a combination of Partitioned Global Address Space (PGAS) languages and libraries, providing one-sided communication paradigm with put, get and accumulate primitives. To support the PGAS models, it is critical to design power efficient and high performance one-sided communication runtime system. In this paper, we design and implement PASCoL, a high performance power aware one-sided communication library using Aggregate Remote Memory Copy Interface (ARMCI), the communication runtime system of Global Arrays. For various communication primitives provided by ARMCI, we study the impact of Dynamic Voltage/Frequency Scaling (DVFS) and a combination of interrupt (blocking)/polling based mechanisms provided by most modern interconnects. We implement our design and evaluate it with synthetic benchmarks using an InfiniBand cluster. Our results indicate that PASCoL can achieve significant reduction in energy consumed per byte transfer without additional penalty for various one-sided communication primitives and various message sizes and data transfer patterns.

99 GENERAL AND MISCELLANEOUS↗

Probabilistic Communication and I/O Tracing with Deterministic Replay at Scale

With today's petascale supercomputers, applications often exhibit low efficiency, such as poor communication and I/O performance, that can be diagnosed by analysis tools. However, these tools either produce extremely large trace files that complicate performance analysis, or sacrifice accuracy to collect high-level statistical information using crude averaging. This work contributes Scala-H-Trace, which features more aggressive trace compression than any previous approach, particularly for applications that do not show strict regularity in SPMD behavior. Scala-H-Trace uses histograms expressing the probabilistic distribution of arbitrary communication and I/O parameters to capture variations. Yet, where other tools fail to scale, Scala-H-Trace guarantees trace files of near constant size, even for variable communication and I/O patterns, producing trace files orders of magnitudes smaller than using prior approaches. We demonstrate the ability to collect traces of applications running on thousands of processors with the potential to scale well beyond this level. We further present the first approach to deterministically replay such probabilistic traces (a) without deadlocks and (b) in a manner closely resembling the original applications. Our results show either near constant sized traces or only sub-linear increases in trace file sizes irrespective of the number of nodes utilized. Even with the aggressively compressed histogram-based traces, our replay times are within 12% to 15% of the runtime of original codes. Such concise traces resembling the behavior of production-style codes closely and our approach of deterministic replay of probabilistic traces are without precedence.

99 GENERAL AND MISCELLANEOUS↗

Parallel implementation of an algorithm for Delaunay triangulation

This work concerns the theory and practice of implementing Tanemura's algorithm for 3D Delaunay triangulation on Intel's Gamma prototype, a 128 processor MIMD computer. Tanemura's algorithm does not vectorize to any significant degree and requires indirect addressing. Efficient implementation on a conventional, vector processing, supercomputer is problematic. Efficient implementation on a parallel architecture is possible, however. In this work, speeds in excess of 8 times a single processor Cray Y-mp are realized on 128 processors of the Intel Gamma prototype.

Merriam, Marshall L.↗

1993 Gordon Bell Prize Winners

The Gordon Bell Prize recognizes significant achievements in the application of supercomputers to scientific and engineering problems. In 1993, finalists were named for work in three categories: (1) Performance, which recognizes those who solved a real problem in the quickest elapsed time. (2) Price/performance, which encourages the development of cost-effective supercomputing. (3) Compiler-generated speedup, which measures how well compiler writers are facilitating the programming of parallel processors. The winners were announced November 17 at the Supercomputing 93 conference in Portland, Oregon. Gordon Bell, an independent consultant in Los Altos, California, is sponsoring $2,000 in prizes each year for 10 years to promote practical parallel processing research. This is the sixth year of the prize, which Computer administers. Something unprecedented in Gordon Bell Prize competition occurred this year: A computer manufacturer was singled out for recognition. Nine entries reporting results obtained on the Cray C90 were received, seven of the submissions orchestrated by Cray Research. Although none of these entries showed sufficiently high performance to win outright, the judges were impressed by the breadth of applications that ran well on this machine, all nine running at more than a third of the peak performance of the machine.

Karp, Alan H.↗

Modeling of wave processes in blocky media with porous and fluid-saturated interlayers

The wave processes in blocky media are analyzed by applying different mathematical models, wherein the elastic blocks interact with each other via pliant interlayers with the complex mechanical properties. Four versions of constitutive equations are considered. In the first version, an elastic interaction between the blocks is simulated within the framework of linear elasticity theory, and the model of elastic–plastic interlayers is constructed to take into account the appearance of irreversible deformation of interlayers at short time intervals. In the second one, the effects of viscoelastic shear in the interblock interlayers are taken into the consideration using the Poynting–Thomson rheological scheme. In the third option, the model of an elastic porous material is used in the interlayers, where the pores collapse if an abrupt compressive stress is applied. In the fourth case, the model of a fluid-saturated material with open pores is examined based on Biot's equations. The collapse of pores is modeled by the generalized rheological approach, wherein the mechanical properties of a material are simulated using four rheological elements. Three of them are the traditional elastic, viscous and plastic elements, the fourth element is the so-called rigid contact, which is used to describe the behavior of materials with the different resistance to tension and compression. It was shown that the thermodynamically consistent model is provided, which means that the energy balance equation is fulfilled for an entire blocky structure, where the kinetic and potential energy of the system is the sum of the kinetic and potential energies of the blocks and interlayers. Under numerical implementation of the interlayers models, the dissipationless finite difference Ivanov's method was used. The splitting method by spatial variables in the combination with the Godunov gap decay scheme was applied in the blocks. As a result, robust and stable computational algorithms are built and tested. Using MPI technology, the parallel software was designed for the modeling of wave processes in 2D setting. The numerical results are presented, discussed and future studies are outlined. - Highlights: • We model the propagation of waves in a blocky medium with compliant interlayers. • Models of porous and fluid saturated interlayers are proposed using Biot's approach. • A parallel computational algorithm is developed for supercomputers of cluster type. • Problems of instant rotation of central and boundary blocks are solved numerically. • Wave motions are compared in the cases of elastic and fluid-saturated interlayers.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Effective Ion Mobility Calculations for Macromolecules by Scattering off Electron Clouds

Broad commercialization and increasing resolving power of ion mobility spectrometry/mass spectrometry (IMS/MS) platforms has engendered an explosion of IMS applications to structural characterization of gas-phase biomolecules. That has renewed interest in more accurate and rapid ion mobility calculations needed to elicit ion geometries from the measurements. An approach based on scattering off electron density isosurfaces (SEDI) that mirrors the physics of molecular collisions was proven superior to the common methods involving atomic coordinates a decade ago, but has remained impractical for large ions because of extreme computational demands. Here, we accelerate SEDI by up to ~500 times using the fragment molecular orbital (FMO) approach for surface generation and the multiplexed scattering algorithm in conjunction with the new grid extrapolation procedure for cross section evaluations. Parallelization of the code on a supercomputer has produced major further speed gains, allowing SEDI calculations for proteins (defined by over a million surface points) with the precision of <0.1% in one minute. Initial tests reveal the anticipated dependences of mobility on the ion charge state and lower cross sections in view of reduced surface roughness. Present developments are expected to lead to broad application of SEDI in IMS studies of macromolecules, enabling more accurate and reliable structural assignments.

Alexeev, Yuri↗

Fast Fault Injection and Sensitivity Analysis for Collective Communications, In: 2015 IEEE International Conference on Cluster Computing

The collective communication operations, which are widely used in parallel applications for global communication and synchronization are critical for application's performance and scalability. However, how faulty collective communications impact the application and how errors propagate between the application processes is largely unexplored. One of the critical reasons for this situation is the lack of fast evaluation method to investigate the impacts of faulty collective operations. The traditional random fault injection methods relying on a large amount of fault injection tests to ensure statistical significance require a significant amount of resources and time. These methods result in prohibitive evaluation cost when applied to the collectives. In this paper, we introduce a novel tool named Fast Fault Injection and Sensitivity Analysis Tool (FastFIT) to conduct fast fault injection and characterize the application sensitivity to faulty collectives. The tool achieves fast exploration by reducing the exploration space and predicting the application sensitivity using Machine Learning (ML) techniques. A basis for these techniques are implicit correlations between MPI semantics, application context, critical application features, and application responses to faulty collective communications. The experimental results show that our approach reduces the fault injection points and tests by 97% for representative benchmarks (NAS Parallel Benchmarks (NPB)) and a realistic application (Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS)) on a production supercomputer. Further, we statistically generalize the application sensitivity to faulty collective communications for these workloads, and present correlation between application features and the sensitivity.

Computer Science↗

Hashing Strategies for the Cray XMT

Several strategies for hashing on a Cray XMT are experimentally compared. These strategies are designed to minimize memory contention. Results show good scalability and indicate under what circumstances each of the strategies should be used.

97 MATHEMATICS AND COMPUTING↗

User Manual - HydraGNN v5.0: Distributed Implementation of Multi-Tasking Graph Neural Networks

This document serves as the user manual for HydraGNN v5.0, a scalable graph neural network (GNN) architecture for simultaneous prediction of multiple target properties using multi-task learning (MTL). This version of HydraGNN has been developed primarily to support the development, training, and deployment of predictive graph-based deep learning (DL) models for atomistic materials modeling. HydraGNN is templated over 13 message-passing policies, including invariant models (GIN, PNA, PNAPlus, GAT, MFC, CGCNN, SAGE, SchNet, DimeNet) and equivariant models (EGNN, PNAEq, PAINN, MACE), and supports distributed training via distributed data parallelism (DDP), DeepSpeed, and Fully Sharded Data Parallelism (FSDP) on leadership-class supercomputers. Although HydraGNN can be applied to problems beyond atomistic materials modeling, its current use is confined to homogeneous graphs. Additional capabilities include machine-learned interatomic potentials with energy-conserving forces, General, Powerful, and Scalable Graph Transformer (GraphGPS) global attention, periodic boundary conditions, hyperparameter optimization, mixed-precision training, and uncertainty quantification.

97 MATHEMATICS AND COMPUTING↗

Viskores: Integrating Parallel Scientific Visualization Research into Applications

Viskores is a scientific visualization library that is the primary deployment of such algorithms to the parallel accelerated processors of modern DOE supercomputers. In this paper, we review the capabilities provided by Viskores and how these capabilities are leveraged by other software in the high-performance computing ecosystem. We discuss the Viskores data representation and pay particular attention to array management. Through this array management we describe how data is adapted between Viskores and other software along with strategies for converting dynamic, polymorphic objects to static representations better suited to GPU processing. We conclude with several examples of Viskores integrating with high-performance software that is used in production today.

Moreland, Ken [ORNL] (ORCID:0000000270513288)↗

Scale-up Unlearnable Examples Learning with High-performance Computing

Recent advancements in AI models, like ChatGPT, are structured to retain user interactions, which could inadvertently include sensitive healthcare data. In the healthcare field, particularly when radiologists use AI-driven diagnostic tools hosted on online platforms, there is a risk that medical imaging data may be repurposed for future AI training without explicit consent, spotlighting critical privacy and intellectual property concerns around healthcare data usage. Addressing these privacy challenges, a novel approach known as Unlearnable Examples (UEs) has been introduced, aiming to make data unlearnable to deep learning models. A prominent method within this area, called Unlearnable Clustering (UC), has shown improved UE performance with larger batch sizes but was previously limited by computational resources (e.g., a single workstation). To push the boundaries of UE performance with theoretically unlimited resources, we scaled up UC learning across various datasets using Distributed Data Parallel (DDP) training on the Summit supercomputer. Our goal was to examine UE efficacy at high-performance computing (HPC) levels to prevent unauthorized learning and enhance data security, particularly exploring the impact of batch size on UE’s unlearnability. Utilizing the robust computational capabilities of the Summit, extensive experiments were conducted on diverse datasets such as Pets, MedMNist, Flowers, and Flowers102. Our findings reveal that both overly large and overly small batch sizes can lead to performance instability and affect accuracy. However, the relationship between batch size and unlearnability varied across datasets, highlighting the necessity for tailored batch size strategies to achieve optimal data protection. The use of Summit’s high-performance GPUs, along with the efficiency of the DDP framework, facilitated rapid updates of model parameters and consistent training across nodes. Our results underscore the critical role of selecting appropriate batch sizes based on the specific characteristics of each dataset to prevent learning and ensure data security in deep learning applications. The source code is publicly available at https: // github. com/ hrlblab/ UE_ HPC .

Zhu, Yanfan [Vanderbilt University, Nashville, TN,↗

The NAS Parallel Benchmarks 2.1 Results

We present performance results for version 2.1 of the NAS Parallel Benchmarks (NPB) on the following architectures: IBM SP2/66 MHz; SGI Power Challenge Array/90 MHz; Cray Research T3D; and Intel Paragon. The NAS Parallel Benchmarks are a widely-recognized suite of benchmarks originally designed to compare the performance of highly parallel computers with that of traditional supercomputers.

Saphir, William↗

Modeling of Radiative Transfer in Protostellar Disks

This program implements a spectral line, radiative transfer tool for interpreting Spitzer Space Telescope observations by matching them with models of protostellar disks for improved understanding of planet and star formation. The Spitzer Space Telescope detects gas phase molecules in the infrared spectra of protostellar disks, with spectral lines carrying information on the chemical composition of the material from which planets form. Input to the software includes chemical models developed at JPL. The products are synthetic images and spectra for comparison with Spitzer measurements. Radiative transfer in a protostellar disk is primarily affected by absorption and emission processes in the dust and in molecular gases such as H2, CO, and HCO. The magnitude of the optical absorption and emission is determined by the population of the electronic, vibrational, and rotational energy levels. The population of the molecular level is in turn determined by the intensity of the radiation field. Therefore, the intensity of the radiation field and the population of the molecular levels are inter-dependent quantities. To meet the computational challenges of solving for the coupled radiation field and electronic level populations in disks having wide ranges of optical depths and spatial scales, the tool runs in parallel on the JPL Dell Cluster supercomputer with C++ and Fortran compiler with a Message Passing Interface. Because this software has been developed on a distributed computing platform, the modeling of systems previously beyond the reach of available computational resources is possible.

VonAllmen, Paul↗

Lightforce Photon-Pressure Collision Avoidance: Efficiency Analysis in the Current Debris Environment and Long-Term Simulation Perspective

This work provides an efficiency analysis of the LightForce space debris collision avoidance scheme in the current debris environment and describes a simulation approach to assess its impact on the long-term evolution of the space debris environment. LightForce aims to provide just-in-time collision avoidance by utilizing photon pressure from ground-based industrial lasers. These ground stations impart minimal accelerations to increase the miss distance for a predicted conjunction between two objects. In the first part of this paper we will present research that investigates the short-term effect of a few systems consisting of 20-kilowatt-class lasers directed by 1.5-meter-diameter telescopes using adaptive optics. The results found such a network of ground stations to mitigate more than 85 percent of conjunctions and could lower the expected number of collisions in Low Earth Orbit (LEO) by an order of magnitude. While these are impressive numbers that indicate LightForce's utility in the short-term, the remaining 15 percent of possible collisions contain (among others) conjunctions between two massive objects that would add large amount of debris if they collide. Still, conjunctions between massive objects and smaller objects can be mitigated. Hence, we choose to expand the capabilities of the simulation software to investigate the overall effect of a network of LightForce stations on the long-term debris evolution. In the second part of this paper, we will present the planned simulation approach for that effort. For the efficiency analysis of collision avoidance in the current debris environment, we utilize a simulation approach that uses the entire Two Line Element (TLE) catalog in LEO for a given day as initial input. These objects are propagated for one year and an all-on-all conjunction analysis is performed. For conjunctions that fall below a range threshold, we calculate the probability of collision and record those values. To assess efficiency, we compare a baseline (without collision avoidance) conjunction analysis with an analysis where LightForce is active. Using that approach, we take into account that collision avoidance maneuvers could have effects on third objects. Performing all-on-all conjunction analyses for extended period of time requires significant computer resources; hence we implemented this simulation utilizing a highly parallel approach on the NASA Pleiades supercomputer.

space debris mitigation↗

Dynamic Mode Decomposition of Unsteady Pressure-Sensitive Paint Measurements for the NASA Unitary Plan Wind Tunnel Tests

This paper discusses the Dynamic Mode Decomposition (DMD) of the Unsteady Pressure-Sensitive Paint (uPSP) measurements, which were collected with four Phantom high-speed cameras at a constant sample frequency in the Ascent Transient Aerodynamics Test (ATAT) of the Space Launch System (SLS) Block 1 cargo vehicle with the Unitary Plan Wind Tunnel (UPWT) 11-by-11-foot Transonic Wind Tunnel in September 2019 at NASA Ames Research Center. The conventional DMD algorithm is based on the Singular Value Decomposition (SVD). For the data with zero mean, the DMD is equivalent to the Discrete Fourier Transform (DFT). Since the uPSP is mainly used to determine the unsteady property of the aerodynamic flow, the DMD of the uPSP measurements is implemented in two steps: (1) subtract the mean value from the uPSP measurement; (2) apply the Fast Fourier Transform (FFT) on the resulting data with zero mean. The DMD of the uPSP measurements with FFT has two advantages: (1) the FFT algorithm is well known for its computational efficiency, therefore, compared to the SVD-based DMD algorithm, the DMD with FFT reduces the computation time; (2) the DMD with FFT can be easily implemented in parallel processing. The DMD outputs were generated with the execution in parallel of a code in C, with libraries of FFTW for FFT and MPI/OpenMP for parallel processing, on the NASA Pleiades supercomputer. In this paper, the results of DMD of the uPSP measurements in the tests of Mach sweep runs of the SLS ATAT are presented, and the effectiveness of the DMD of the uPSP measurements in the diagnosis of the unsteady, aerodynamic phenomena is demonstrated. The work described in this paper is a part of NASA’s development of a new state-of-the-art uPSP capability in production wind tunnels. Funding for this research was provided by the NASA Aeroscience Evaluation and Test Capabilities Project.

Pressure-Sensitive Paint↗

Spectral Proper Orthogonal Decomposition of uPSP Measurements in Recent NASA Ames Wind Tunnel Test

This paper discusses Spectral Proper Orthogonal Decomposition (SPOD) of the Unsteady Pressure-Sensitive Paint (uPSP) measurements in recent NASA Ames wind tunnel test. The uPSP measurements were collected using Innovative Scientific Solutions, Inc. (ISSI) porous, fast-response pressure-sensitive paint, 40 ISSI four-inch air-cooled Light-Emitting Diodes, and 8 Phantom v2512 high-speed cameras at 10,000 frames per second in the uPSP Launch Vehicle Demonstration Test (LVDT) of the Space Launch System (SLS) vehicle in the 11-by 11-foot transonic test section of the Unitary Plan Wind Tunnel at NASA Ames Research Center in April 2024. SPOD is derived from a space-time proper orthogonal decomposition problem for statistically stationary flows. SPOD modes are determined in the frequency domain. Each SPOD mode oscillates at a single frequency. SPOD can be viewed as an extension of the Discrete Fourier Transform composition and the Dynamic Mode Decomposition. In this paper, the outputs of SPOD of the uPSP measurements in the uPSP LVDT are presented and the effectiveness of SPOD in the identification, diagnosis and analysis of the aerodynamic and aeroacoustic phenomena is demonstrated. The unsteady and dynamic property of the pressure field on the surface of the SLS Block 1B crew vehicle is presented with the visualization of the SPOD modes of the uPSP measurements in the tests of a Mach sweep run of the uPSP LVDT. The SPOD outputs were generated with the execution in parallel of a code in Python, with the library of Message Passing Interface for parallel processing, on the NASA Pleiades supercomputer. The work described in this paper is a part of NASA’s development of a new state-of-the-art uPSP capability in production wind tunnels. Funding was provided by the NASA Aerosciences Evaluation and Test Capabilities Portfolio Office.

Aeroacoustics↗