Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “HPC performance modeling”

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 505 records · Page 28

Comparative Study of Large Language Model Architectures on Frontier

Large language models (LLMs) have garnered significant attention in both the AI community and beyond. Among these, the Generative Pre-trained Transformer (GPT) has emerged as the dominant architecture, spawning numerous variants. However, these variants have undergone pre-training under diverse conditions, including variations in input data, data preprocessing, and training methodologies, resulting in a lack of controlled comparative studies. Here we meticulously examine two prominent open-sourced GPT architectures, GPT-NeoX and LLaMA, leveraging the computational power of Frontier, the world’s first Exascale supercomputer. Employing the same materials science text corpus and a comprehensive end-to-end pipeline, we conduct a comparative analysis of their training and downstream performance. Our efforts culminate in achieving state-of-the-art performance on a challenging materials science benchmark. Furthermore, we investigate the computation and energy efficiency, and propose a computationally efficient method for architecture design. To our knowledge, these pre-trained models represent the largest available for materials science. Our findings provide practical guidance for building LLMs on HPC platforms.

Yin, Junqi↗

SSAPy - Space Situational Awareness for Python

SSAPy is a fast and flexible orbit modeling and analysis tool for orbits spanning from low-Earth into the cislunar regime. Orbits can be flexibly specified from common input formats such as Keplerian elements or two-line element (TLE) data files. SSAPy allows users to model satellites and specify parameters such as satellite area, mass, and drag coefficients. SSAPy includes a customizable force-propagation with a range of Earth, Lunar, radiation, atmospheric, and maneuvering models. SSAPy makes use of various community integration methods and can calculate time-evolved orbital quantities, including satellite magnitudes and state vectors. Users can specify various space- and ground-based observation models with support for multiple coordinate and reference frames. SSAPy also supports orbit analysis and propagation methods such as multiple hypothesis tracking and has built-in uncertainty quantification. The majority of SSAPy’s methods are vectorized and parallelizable, allowing for effective use of high-performance computer (HPC) systems. Finally, SSAPy has plotting functionality, allowing users to visualize orbits and trajectories. Examples are shown in Figure 1 and Figure 2.

97 MATHEMATICS AND COMPUTING↗

The Institute for Sustained Performance, Energy, and Resilience Final Project Report

Over the course of the five years of the Scientific Discovery through Advanced Computation (SciDAC-3) program, computational scientists working on behalf of the Department of Energy's Office of Science (DOE SC) exploited a new generation of petascale computing resources to make previously inaccessible discoveries in a broad range of disciplines including physics, chemistry, and materials science. The computing systems underpinning this work increased in performance potential from tens to hundreds of PFlop/s, but in the process evolving significantly from architectures in 2011, when SciDAC-3 began. Although Moore's law continues unabated, the end of Dennard scaling has necessitated a fundamental shift in computer architecture focused on power efficiency. To that end, processors are increasingly varied as they strive to satisfy performance, productivity, reliability, and energy efficiency in the face of divergent computational requirements. The diversity among these machines presents a number of challenges to merely porting today's scientific applications, much less achieving good performance. Therefore, at the beginning of SciDAC-3, USC and its collaborators anticipated that both vastly increased scale (e.g., more chips, more cores per chip, wider SIMD) and heterogeneity would exacerbate performance optimization challenges while simultaneously promoting the issues of energy and heterogeneity will exacerbate performance optimization challenges while simultaneously promoting the issues of energy consumption and resilience to the forefront. To ensure that DOE's computational scientists could successfully exploit emerging high performance computing (HPC) systems, DOE SC’s Advanced Scientific Computing Research Program (ASCR) funded the SciDAC-3 Institute for Sustained Performance, Energy, and Resilience (SUPER). Led by USC, this was a broadly based project with expertise in compilers and system tools, performance engineering, energy management, and resilience. SUPER followed the successful model developed in the previous SciDAC-2 Performance Engineering Research Institute (PERI) of leveraging the research investments DOE and others have made and integrating the results to create new capabilities beyond the reach of any one group. To accomplish this, SUPER was organized into multi-institution teams conducting the research activities in performance portability, energy efficiency, exascale resilience, and overall optimization. SUPER researchers also collaborated with other SciDAC-3 institutes, 13 of the 19 Scientific Computation Application Partnerships (SAP), as well as the broader DOE computational sciences community. These collaborations both focused our research on the real challenges facing petascale scientific computing as well as ensured the broad and immediate impact of the results of our research.

97 MATHEMATICS AND COMPUTING↗

GASNet-EX Memory Kinds: Support for Device Memory in PGAS Programming Models

There is an emerging need for adaptive, lightweight communication in irregular HPC applications at exascale, where GPU accelerators provide the majority of available compute cycles. To address this need, Lawrence Berkeley National Lab is developing a programming system to support distributed-memory HPC application development using the Partitioned Global Address Space (PGAS) model. This work includes two major components: UPC++ and GASNet-EX. UPC++ is a C++ template library providing Remote Memory Access (RMA) and Remote Procedure Call (RPC) communication interfaces. GASNet-EX is a portable, high-performance communication middleware library, used by the implementations of UPC++ and many other PGAS programming models. We describe recent advances in GASNet-EX to efficiently implement zero-copy Remote Memory Access (RMA) communication to and from memory on accelerator devices such as GPUs. We demonstrate performance improvements via benchmark results from UPC++ (on Summit) and the Legion programming system (on DGX-1), both using GASNet-EX for communication.

Hargrove, Paul H↗

Performance-Aligned LLMs for Generating Fast HPC Code

Optimizing scientific software is a difficult task because codebases are often large and complex, and performance can depend upon several factors including the algorithm, its implementation, and hardware among others. Causes of poor performance can originate from disparate sources and be difficult to diagnose. Recent years have seen a multitude of work that use large language models (LLMs) to assist in software development tasks. However, these tools are trained to model the distribution of code as text, and are not specifically designed to understand performance aspects of code. In this work, we introduce a reinforcement learning based methodology to align the outputs of code LLMs with performance. This allows us to build upon the current code modeling capabilities of LLMs and extend them to generate better performing code. Here, we demonstrate that our fine-tuned model improves the expected speedup of generated code over base models for a set of benchmark tasks from 0.9 to 1.6 for serial code and 1.9 to 4.5 for OpenMP parallel code.

Computer science↗

Understanding I/O workload characteristics of a Peta-scale storage system

Understanding workload characteristics is critical for optimizing and improving the performance of current systems and software, and architecting new storage systems based on observed workload patterns. In this paper, we characterize the I/O workloads of scientific applications of one of the world's fastest high performance computing (HPC) storage cluster, Spider, at the Oak Ridge Leadership Computing Facility (OLCF). OLCF flagship petascale simulation platform, Titan, and other large HPC clusters, in total over 250 thousands compute cores, depend on Spider for their I/O needs. We characterize the system utilization, the demands of reads and writes, idle time, storage space utilization, and the distribution of read requests to write requests for the Peta-scale Storage Systems. From this study, we develop synthesized workloads, and we show that the read and write I/O bandwidth usage as well as the inter-arrival time of requests can be modeled as a Pareto distribution. We also study the I/O load imbalance problems using I/O performance data collected from the Spider storage system.

I/O↗

Distributed-Memory Sparse Deep Neural Network Inference Using Global Arrays

Partitioned Global Address Space (PGAS) models exhibit tremendous promise in developing efficient and productive distributed-memory parallel applications. They have been used extensively in scientific computations due to conveniently offering a ``shared-memory''-like model and convenient interfaces that separate communication with synchronization. Traditionally, PGAS communication models have been applied to dense/contiguously distributed data, but most modern applications depict varied levels of sparsity. Existing PGAS models require certain adaptations to support distributed sparse computations, since associated computations often require matrix arithmetic, in addition to data movement. The Global Arrays toolkit from Pacific Northwest National Laboratory (PNNL) is one of the earliest PGAS models to combine one-sided data communication and distributed matrix operations and is still used in the popular NWChem quantum chemistry suite. Recently, we have expanded the Global Arrays toolkit to support common sparse operations, like sparse matrix-dense matrix multiplies (SpMM), sparse matrix-sparse matrix multiplication (SpGEMM) and Sampled Dense-Dense Matrix Multiplication (SDDMM). As it turns out, these operations are the bedrock of sparse Deep Learning (DL); sparse deep neural networks and Graph Neural Networks (GNNs) have gained increasing attention recently in achieving speedups on training and inference with reduced memory footprints. Unlike scientific applications in High Performance Computing (HPC), modern (distributed-memory capable) DL toolkits often rely on non-standardized and closed-source vendor software optimizations, creating challenges in software-hardware co-design at scale. Our goal is to support a variety of distributed-memory sparse matrix operations and helper functions in the newly created Sparse Global Arrays (SGA), such that it is possible to build portable and productive Machine Learning scenarios for algorithm/software and hardware codesign purposes. Contemporary data-parallel schemes for training/inference are undergoing a major overhaul since model replication limits scalability and causes resource inefficiencies. As such, we have adopted tensor parallelism in decomposing the model and inputs, to mitigate memory issues. Current implementation is built on top of MPI and uses CPUs to maximize the portability across the platforms.

Distributed computing, machine learning↗

Develop a new integrated macro→micro←nano (MMN) multiscale modeling framework to optimize high strength aluminum alloys and processes for vehicle light-weighting​

Bending tests provide a means to study plane strain fracture performance of 6000 series aluminum alloys. Metrics from bending tests have been correlated with self-pierce riveting (SPR) performance of a high strength AA6111 automotive aluminum alloy in previous works. Using the ORNL HPC resources, this project developed an innovative macro→micro←nano (MMN) multiscale microstructure-based finite element (FE) code to further the understanding of microstructural relationship to fracture properties of high strength 6000 series alloys.

36 MATERIALS SCIENCE↗

Real-time measurements of temperature, pressure and moisture profiles in High-Performance Concrete exposed to high temperatures during neutron radiography imaging

High-Performance Concrete (HPC) is particularly prone to explosive spalling when exposed to high temperature. Although the exact causes that lead to spalling are still being debated, moisture transport during heating plays an important role in all proposed mechanisms. In this study, slabs made of high-performance, low water-to-binder ratio mortars with addition of superabsorbent polymers (SAP) and polypropylene fibers (PP) were heated from one side on a temperature-controlled plate up to 550 °C. A combination of measurements was performed simultaneously on the same sample: moisture profiles via neutron radiography, temperature profiles with embedded thermocouples and pore pressure evolution with embedded pressure sensors. Spalling occurred in the sample with SAP, where sharp profiles of moisture and temperature were observed. No spalling occurred when PP-fibers were introduced in addition to SAP. The experimental procedure described here is essential for developing and verifying numerical models and studying measures against fire spalling risk in HPC.

42 ENGINEERING↗

Improving Additive Manufactured Component Performance through Multi-Scale Microstructure Simulation and Process Optimization

The purpose of this project was to utilize computational tools to understand the relationships between processing, microstructure, and properties for additively manufactured (AM) aluminum alloys for automotive applications, and to provide an engineering solution for helping to optimize process conditions. The project leverages ORNL developments in computational modeling, including AM process modeling, phase-field based microstructure evolution predictions, and data analytics techniques for mapping process conditions to material outcomes. The project utilized an Al-Cu-Mn-Zr alloy as a model material for studying formation of defects and microstructural features in response to variations in process conditions. Based on both pre-existing experimental data and simulation results, statistical process maps were constructed to identify regions of process space with minimal defect formation and advantageous microstructures and properties. The software tools used for this purpose were successful disseminated to GM, who were able to successful compile the relevant HPC codes within their own computing ecosystem and perform initial calculations to reproduce ORNL results.

36 MATERIALS SCIENCE↗

A Hands-On Curriculum for Training in HPC Cluster Deployment and Management

This paper presents the design, methodology, and outcomes of the High-Performance Computing Technologies (HPCT) course, a hands-on training program focused on the system-side of HPC cluster deployment and administration. Delivered as part of the Master in High Performance Computing (MHPC) program, the course introduces students to key concepts in cluster configuration, including networking, software stack provisioning, job scheduling, and monitoring. Initially taught in person, the course was transitioned to an online format during the COVID-19 pandemic. This shift led to the development of openly available instructional material and a flipped-classroom approach that continues to support both in-person and hybrid delivery. All course materials are publicly available at www.hpc.temple.edu/mhpc/hpc-technology/index.html. By documenting the structure, infrastructure, and evolution of HPCT, this paper offers a model for accessible HPC system training that supports workforce development in computational science.

Posada Correa, Fernando [ORNL] (ORCID:000000022565↗

Distributing Power Grid State Estimation on HPC Clusters A System Architecture Prototype

The future power grid is expected to further expand with highly distributed energy sources and smart loads. The increased size and complexity lead to increased burden on existing computational resources in energy control centers. Thus the need to perform real-time assessment on such systems entails efficient means to distribute centralized functions such as state estimation in the power system. In this paper, we present our early prototype of a system architecture that connects distributed state estimators individually running parallel programs to solve non-linear estimation procedure. The prototype consists of a middleware and data processing toolkits that allows data exchange in the distributed state estimation. We build a test case based on the IEEE 118 bus system and partition the state estimation of the whole system model to available HPC clusters. The measurement from the testbed demonstrates the low overhead of our solution.

Distributed architecture↗

Thermosyphon Cooler Hybrid System for Water Savings in an Energy-Efficient HPC Data Center: Modeling and Installation: Preprint

The Thermosyphon Cooler Hybrid System (TCHS) integrates the control of a dry heat rejection device, the thermosyphon cooler (TSC), with an open cooling tower. A combination of equipment and controls, this new heat rejection system embraces the 'smart use of water,' using evaporative cooling when it is most advantageous and then saving water and modulating toward increased dry sensible cooling as system operations and ambient weather conditions permit. Innovative fan control strategies ensure the most economical balance between water savings and parasitic fan energy. The unique low-pressure-drop design of the TSC allows water to be cooled directly by the TSC evaporator without risk of bursting tubes in subfreezing ambient conditions. Johnson Controls partnered with the National Renewable Energy Laboratory (NREL) and Sandia National Laboratories to deploy the TSC as a test bed at NREL's high-performance computing (HPC) data center in the first half of 2016. Located in NREL's Energy Systems Integration Facility (ESIF), this HPC data center has achieved an annualized average power usage effectiveness rating of 1.06 or better since 2012. Warm-water liquid cooling is used to capture heat generated by computer systems direct to water; that waste heat is either reused as the primary heat source in the ESIF building or rejected using evaporative cooling. This data center is the single largest source of water and power demand on the NREL campus, using about 7,600 m3 (2.0 million gal) of water during the past year with an hourly average IT load of nearly 1 MW (3.4 million Btu/h) -- so dramatically reducing water use while continuing efficient data center operations is of significant interest. Because Sandia's climate is similar to NREL's, this new heat rejection system being deployed at NREL has gained interest at Sandia. Sandia's data centers utilize an hourly average of 8.5 MW (29 million Btu/h) and are also one of the largest consumers of water on Sandia's site. In addition to describing the installation of the TSC and its integration into the ESIF, this paper focuses on the full heat rejection system simulation program used for hourly analysis of the energy and water consumption of the complete system under varying operating scenarios. A follow-up paper will detail the test results. The evaluation of the TSC's performance at NREL will also determine a path forward at Sandia for possible deployment in a large-scale system not only for data center use but also possibly site wide.

97 MATHEMATICS AND COMPUTING↗

Experiences with High-Level Programming Directives for Porting Applications to GPUs. In: Facing the Multicore--Challenge II, Lecture Notes in Computer Science.

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the different node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressivity and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

CUDA Implementation↗

An Efficient Storage-Driven Machine Learning Model for Performance in the Era of Multimodal Scientific Data

Scientific workflows are increasingly relying on machine learning (ML), simulation, and hybrid techniques to predict, understand, and optimize the behavior of complex experiments. High-performance computing has greatly improved researchers’ ability to acquire diverse data modalities in these workflows. Recent studies suggest that the performance of machine learning models can be improved by integrating data from various sources. Unfortunately, these workloads pose unprecedent pressure on the network storage to meet the demands associated with accessing these multimodal data. To mitigate the impact of intensive IO, we propose a solution that utilizes a multi-tier High-Performance Computing (HPC) distributed storage and data processing framework, placing computation where the data resides for better performance. By adopting this project, the scientific community will gain new opportunities to explore multimodal storage-driven possibilities, integrating multiple scientific data sources with advanced streaming frameworks. Additionally, our framework effectively utilizes computing resources and bridges the gaps identified by HPC experts. Our proposed approach tackles scalability and persistence challenges by leveraging native persistency, which has posed difficulties in traditional approaches. Furthermore, we seek to enhance fault-tolerance and load-balance of computations by leveraging real-time streaming in diverse scientific computing environments, thereby propelling advanced scientific computing research into the next generation.

97 MATHEMATICS AND COMPUTING↗

Brochure for the DOE Office of Science Workshop on Envisioning Frontiers in AI and Computing for Biological Research

In February of 2025 a joint ASCR/BER workshop was held to identify key transformational research directions for understanding biology using artificial intelligence (AI), digital twins and high-performance (HPC) computational methods to facilitate scientific discovery and innovation in support of the Department of Energy mission. AI technologies offer exciting new groundbreaking methods to analyze large volumes of complex biological data, thereby greatly accelerating the ability to understand, predict, and design biological processes for beneficial purposes. In the laboratory, the bridging of AI-enabled automated experimental technologies, HPC and digital twins will provide potent tools for researchers to explore the fundamental nature of biology and harness its inherent metabolic potential for a variety of beneficial purposes. The focus of this workshop was on how high-performance computational methods can impact this objective by exploring digital twins, foundational models, and data-driven approaches with applications to advance automated laboratory experiments, modeling of complex living systems and engineering new functions into plants and microbial systems relevant to DOE mission. Workshop attendees with expertise in plant science, microbiology, mathematics, computer science, and AI assessed the current state of the science, trends, and AI challenges at the interface of plant and microbial systems biology and computational science to identify opportunities for high-impact research. This collaborative effort capitalized on ASCR's advancements in applied mathematics, computer science, and Exascale systems, and BER's expertise in basic genomics-enabled research on DOE relevant plant and microbial systems. The workshop culminated in four key priority research directions to guide future research and development within DOE Office of Science programs.

59 BASIC BIOLOGICAL SCIENCES↗

Experiences with High-Level Programming Directives for Porting Applications to GPUs

HPC systems now exploit GPUs within their compute nodes to accelerate program performance. As a result, high-end application development has become extremely complex at the node level. In addition to restructuring the node code to exploit the cores and specialized devices, the programmer may need to choose a programming model such as OpenMP or CPU threads in conjunction with an accelerator programming model to share and manage the difference node resources. This comes at a time when programmer productivity and the ability to produce portable code has been recognized as a major concern. In order to offset the high development cost of creating CUDA or OpenCL kernels, directives have been proposed for programming accelerator devices, but their implications are not well known. In this paper, we evaluate the state of the art accelerator directives to program several applications kernels, explore transformations to achieve good performance, and examine the expressiveness and performance penalty of using high-level directives versus CUDA. We also compare our results to OpenMP implementations to understand the benefits of running the kernels in the accelerator versus CPU cores.

43 PARTICLE ACCELERATORS↗

Improving Thermal Management Strategies for Data Centers: A Physical Testbed Incorporating Small Modular Reactor and Microreactor Technology

This study aims to accelerate the demonstration of various thermal management systems for data centers using nuclear-generated heat to enhance energy and grid reliability. Utilizing mobile containerized and stationary test beds at INL's High Performance Computing (HPC) facility, this project integrates with various nuclear-related energy systems testing facilities. Key components include immersion cooling apparatus, absorption chillers, and adjustable thermal management simulators. Tasks involve acquiring necessary hardware, sensors, and cooling apparatus, engaging with data center industry stakeholders, and providing a testing platform for algorithms, models, tools, and software. The objective is to expedite the deployment of nuclear-powered data centers, thereby improving energy reliability and affordability.

21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN↗