Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “High Performance Computing (HPC)”

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 595 records · Page 33
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Performance Study of CXL Memory Topology

This paper presents a comprehensive evaluation of the performance impact of various Compute Express Link (CXL) memory topologies, with a particular emphasis on CXL switches, in the context of High- Performance Computing (HPC) and Large Language Model (LLM) inference workloads. Our study unveils significant performance variations across different topologies, demonstrating that certain configurations yield superior performance for specific workloads. These findings underscore the critical importance of tailored topol- ogy selection in optimizing system performance. Additionally, we address the inherent challenges associated with integrating CXL switches, including overhead considerations and routing complex- ities. Our research highlights the necessity for thorough evalua- tion methodologies to fully leverage CXL technology’s potential in contemporary computing environments. These insights provide valuable guidance for system architects and data center operators in designing and optimizing CXL-based infrastructures for diverse workload requirements.

CXL, memory, Artificial Intelligence (AI), HPC↗
DOE OSTI2025

Secure API-Driven Research Automation to Accelerate Scientific Discovery

The Secure Scientific Service Mesh (S3M) provides API-driven infrastructure to accelerate scientific discovery through automated research workflows. By integrating near real-time streaming capabilities, intelligent workflow orchestration, and fine-grained authorization within a service mesh architecture, S3M enables secure and flexible programmatic access to high performance computing (HPC) resources. This framework allows intelligent agents and experimental facilities to dynamically provision resources and execute complex workflows, accelerating experimental lifecycles, and enabling AI-augmented autonomous science. S3M establishes a modern foundation for scientific computing infrastructure that significantly reduces traditional barriers between researchers, computational resources, and experimental facilities.

Skluzacek, Tyler [ORNL] (ORCID:0000000322424931)↗
DOE OSTI2025

FAIR Ecosystems for Science at Scale

High Performance Computing (HPC) centers provide resources to users who require greater scale to “get science done”. They deploy infrastructure with singular hardware architectures, cutting-edge software environments, and stricter security measures as compared with users’ own resources. As a result, users often create and configure digital artifacts in ways that are specialized for the unique infrastructure at a given HPC center. Each user of that center will face similar challenges as they develop specialized solutions to take full advantages of the center’s resources, potentially resulting in significant duplication of effort. Much duplicated effort could be avoided, however, if users of these centers found it easier to discover others’ solutions and artifacts as well as share their own. The FAIR principles address this problem by presenting guidelines focused around metadata practices to be implemented by vaguely defined “communities”; in practice, these tend to gather by domain (e.g. bioinformatics, geosciences, agriculture). Domain-based communities can unfortunately end up functioning as silos that tend both to inhibit sharing of solutions and best practices as well as to encourage fragile and unsustainable improvised solutions in the absence of best-practice guidance. We propose that these communities pursuing “science at scale” be nurtured both individually and collectively by HPC centers so that users can take advantage of shared challenges across disciplines and potentially across HPC centers. We describe an architecture based on the EOSC-Life FAIR Workflows Collaboratory, specialized for use with and inside HPC centers such as the Oak Ridge Leadership Computing Facility (OLCF), and we speculate on user incentives to encourage adoption. We note that a focus on FAIR workflow components rather than FAIR workflows is more likely to benefit the users of HPC centers.

Wilkinson, Sean [ORNL] (ORCID:0000000214437479)↗
DOE OSTI2025

Characterizing the Impact of GPU Power Management on an Exascale System

As GPU-accelerated high-performance computing (HPC) systems approach exascale performance, controlling energy consumption without compromising throughput is essential. Architectures such as the AMD MI250X-based Frontier supercomputer provide runtime mechanisms like frequency and power capping, enabling energy tuning without modifying application code. Although both target energy reduction, they operate via distinct hardware control paths and influence workloads differently. We present a comprehensive evaluation of these strategies on a leadership-class system using diverse HPC proxy applications representative of production workloads. Our study analyzes performance–energy trade-offs across multiple capping levels, node counts (1 and 32), and application profiles. Results show that frequency capping generally achieves higher energy efficiency and scalability, with gains of up to 13.2% without performance loss, while power capping is more effective for single-node runs or bursty GPU utilization. We also provide practical guidelines to help system administrators and users balance energy efficiency and performance in large-scale scientific workloads.

Costa, Mariana [Universidade Federal do Rio Grande↗
DOE OSTI2025

Data Readiness for Scientific AI at Scale

This paper examines how Data Readiness for AI (DRAI) principles apply to leadership-scale scientific datasets used to train foundation models. We analyze archetypal workflows across four representative domains—climate, nuclear fusion, bio/health, and materials—to identify common preprocessing patterns and domain-specific constraints. We introduce a two-dimensional readiness framework that combines canonical preprocessing patterns with a five-level operational readiness scale, both tailored to high-performance computing (HPC) environments. This framework helps outline key challenges in transforming large-scale scientific data into formats suitable for scalable AI training. Together, these dimensions form a conceptual maturity matrix that characterizes scientific data readiness and guides infrastructure development toward standardized, cross-domain support for scalable and reproducible AI for science.

Brewer, Wes [ORNL] (ORCID:0000000236393956)↗
DOE OSTI2025

IRIS-MASH: Efficient Multi-device Asynchronous Multi-Stream Heterogeneous Computing

In the rapidly evolving field of high-performance computing (HPC), effectively leveraging heterogeneous devices through asynchronous task programming is paramount. This paper presents a robust asynchronous task programming model tailored for a multi-device, multi-stream execution environment that incorporates a diverse array of heterogeneous computing units, including GPUs from various vendors and other accelerators. Current state-of-the-art task programming models provide methodologies to support asynchronous task executions, but they typically handle homogeneous devices using native programming languages, while support for heterogeneous devices is limited to frameworks like OpenCL. This gap presents significant challenges in abstracting heterogeneous devices to harness their true asynchronous capabilities effectively using their native programming languages. By implementing asynchronous task execution, our model significantly boosts the performance of tiled algorithm task graphs through overlapping data transfers with computation and enabling the simultaneous execution of multiple kernels. We integrate this approach into a heterogeneous Intelligent Runtime System (IRIS) and assess its performance using a suite of tiled algorithm benchmarks from the heterogeneous math kernels library (MatRIS) based on IRIS. Experimental results demonstrate a performance improvement ranging from 1.6 × to 2 × over IRIS without asynchronous support, and a notable 22% performance enhancement compared to established runtime systems such as StarPU and PaRSEC. This approach significantly improves computation efficiency of HPC workflows and provides a solid base for future exploration and development in the area of asynchronous task programming in heterogeneous systems.

Miniskar, Narasinga Rao [ORNL] (ORCID:000000018259↗
DOE OSTI2015

Emulating Multiple Inheritance in Fortran 2003/2008

Although the high-performance computing (HPC) community increasingly embraces object-oriented programming (OOP), most HPC OOP projects employ the C++ programming language. Until recently, Fortran programmers interested in mining the benefits of OOP had to emulate OOP in Fortran 90/95. The advent of widespread compiler support for Fortran 2003 now facilitates explicitly constructing object-oriented class hierarchies via inheritance and leveraging related class behaviors such as dynamic polymorphism. Although C++ allows a class to inherit from multiple parent classes, Fortran and several other OOP languages restrict or prohibit explicit multiple inheritance relationships in order to circumvent several pitfalls associated with them. Nonetheless, what appears as an intrinsic feature in one language can be modeled as a user-constructed design pattern in another language. The present paper demonstrates how to apply the facade structural design pattern to support a multiple inheritance class relationship in Fortran 2003. The design unleashes the power of the associated class relationships for modeling complicated data structures yet avoids the ambiguities that plague some multiple inheritance scenarios.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

The ORNL Overhead Vehicle Dataset (OOVD)

Roadways are critical to meeting the mobility and economic needs of the nation. The United States uses 28% of its energy in moving goods and people, with approximately 60% of that utilized by cars, light trucks, and motorcycles. Thus, improved transportation efficiency is vital to America’s economic progress. The increasing congestion and energy resource requirements of transportation systems for metropolitan areas require research in methods to improve and optimize control methods. Coordinating and optimizing traffic in urban areas may introduce hundreds of thousands of vehicles and traffic management systems, which can require high performance computing (HPC) resources to model and manage. This data set was created to understand the potential for machine learning, computer vision, and HPC to improve the energy efficiency aspects of traffic control by leveraging GRIDSMART traffic cameras as sensors for adaptive traffic control, with a sensitivity to the fuel consumption characteristics of the traffic in the camera’s visual field. GRIDSMART cameras—an existing, fielded commercial product—sense the presence of vehicles at intersections and replace more conventional sensors (such as inductive loops) to issue calls to traffic control. These cameras, which have horizon-to-horizon view, offer the potential for an improved view of the traffic environment which can be used to generate better control algorithms.

Karnowski, Thomas P.↗
DOE OSTI2019

Slycat v 2.0

Slycat™ is a web-based system for analysis of large, high-dimensional data such as that produced by High Performance Computing (HPC) platforms. The Slycat™ server integrates data ingestion, scalable analysis, data management, and visualization with commodity web clients using a multi-tiered hierarchy of data and model storage. Analysis inputs are pushed or pulled into Slycat™, where models are generated and analysis artifacts are stored in a project database. These artifacts are the basis for visualizations that are delivered to users’ desktops through ordinary web browsers, eliminating the need to build and deploy platform-specific client applications. Slycat™ currently provides four types of analysis: canonical correlation analysis (CCA) to model relationships between inputs and output metrics, time series analysis featuring clustering and comparative visualization of waveforms, parameter space visualization including remote retrieval of images and movies from HPC disks, and VideoSwarm for modeling video similarities over time.

Letter, Matthew↗
DOE OSTI2013

Gnawts

High Performance Computing (HPC) systems generate millions of inter-related event log lines per day from thousands of computers, providing feedback on both successful and unsuccessful operations within the system. Thorough analysis of these logs is often required for root cause analysis of problems and tuning for greatest performance. This "app" for the "Gnawts" log analysis software (http://www.gnawts.com) facilitates faster and deeper insights into HPC system logs.

Stearley, Jon↗
DOE OSTI2018

Comb

Comb is a communication performance benchmarking tool. It is used to determine performance tradeoffs in implementing communication patterns on high performance computing (HPC) platforms. At its core comb runs combinations of communication patterns with execution patterns, and memory spaces. The current set of capabilities Comb provides includes: o Configurable structured mesh halo exchange communication. o A variety of communication patterns based on grouping messages. o A variety of execution patterns including serial, openmp threading, cuda, cuda batched kernels, and cuda persistent kernels. o A variety of memory spaces including default system allocated memory, pinned host memory, cuda device memory, and cuda managed memory with different cuda memory advice.

Burmark, Jason↗
DOE OSTI2018

Umpire

The Umpire library consists of a collection of software components that provide memory and resource management capabilities that can be used by diverse high performance computing (HPC) applications. The current set of capabilities that Umpire provides Includes: • Flexible allocators to provision memory In the range of available memory resources on a target system. • An extensible allocation strategy model, allowing custom algorithms to be used to allocate memory resources in a scalable and high­performance fashion. • Memory-location aware operations that can be used to migrate and modify memory using specialized functions dependent on source and target data locations • A set of Introspection capabilities to allow the user to query the state of the different allocators and memory allocations.

ECP↗
DOE OSTI2017

RAJA Performance Suite

The RAJA Performance Suite is designed to evaluate performance of the RAJA performance portability library on a wide variety of important high performance computing (HPC) algorithmic lulmels. These kernels assess compiler optimizations and various parallel programming model backends accessible through RAJA, such as OpenMP, CUDA, etc. The Initial version of the suite contains 25 computational kernels, each of which appears in 6 variants: Baseline SequcntiaJ, RAJA SequentiaJ, Baseline OpenMP, RAJA OpenMP, Baseline CUDA, RAJA CUDA. All variants of each kernel perform essentially the same mathematical operations and the loop body code for each kernel is identical across all variants. There are a few kernels, such as those that contain reduction operations, that require CUDA-specific coding for their CUDA variants. ActuaJ computer instructions executed and how they run in parallel differs depending on the parallel programming model backend used and which optimizations are perfonned by the compiler used to build the Perfonnance Suite executable. The Suite will be used primarily by RAJA developers to perform regular assessments of RAJA performance across a range of hardware platforms and compilers as RAJA features are being developed. It will also be used by LLNL hardware and software vendor panners for new defining requirements for future computing platform procurements and acceptance testing. In particular, the RAJA Performance Suite will be used for compiler acceptance testing of the upcoming CORAUSierra machine {initial LLNL delivery expected in late-2017/early 2018) and the CORAL-2 procurement. The Suite will aJso be used to generate concise source code reproducers of compiler and runtime issues we uncover so that we may provide them to relevant vendors to be fixed.

Hornung, RichardD.↗
DOE OSTI2018

ShareAnalytics

ShareAnalytics is a lightweight extensible platform for collaborative data analytics and facility characterization via Non-intrusive Load Monitoring (NILM). It consists of three core components: a full stack web application, a dashboard tool for analyzing streaming data and a High Performance Computing (HPC) cluster leveraging slurm for resource management to perform real time analysis. The entire framework is implemented in Docker for ease of deployment and to allow the user to test the platform before making a significant investment in hardware if they require additional computing resources. While many other platforms for analyzing streaming data exists, ShareAnalytics is unique in that HPC capabilities and facility characterization via scored analytics are fully integrated into the platform.

Oesch, TimothyS↗
DOE OSTI2020

eBPF Community ID Calculation

The code titled “eBPF Community ID Calculation” allows easy correlation of monitoring logs with existing network forensics logs on high-performance computing (HPC) systems. This code implements the Community ID Flow Hashing standard to calculate community ID Flow identifiers. These identifiers will be used to correlate host forensics logs with network forensics logs. The correlation provides valuable insight and the ability to easily pivot between disparate data sources. The computer code removes the libpcap dependency from the code base.

Maerz, StefanA↗
DOE OSTI2020

Highly Asynchronous Visitor Queue Graph Toolkit

HavoqGT (Highly Asynchronous Visitor Queue Graph Toolkit) is a framework for expressing asynchronous vertex-centric graph algorithms, and executing them on High Performance Computing (HPC) systems. It provides a vertex 'visitor' interface, where actions are defined at an individual vertex level, and contains a suite of classic graph algorithms. HavoqGT is capable of processing large graphs stored in NVRAM (SSDs) using a memory mapped interface.

Reza, TahsinA.↗
DOE OSTI2012

Warewulf 3.0

Warewulf is a scalable systems management suite originally developed to manage large high-performance Linux clusters. Focused on general scalable systems management, it includes a framework for system configuration, management, provisioning/installation, monitoring, event notification, and more via a modular plugin architecture. Install the components and features you need or leverage the existing system configuration stored within Warewulf to create custom solutions to meet your performance needs. One of the goals of the project is to grow the project by allowing other organizations and developers to contribute back to the project under an open source BSD license. We have been successful with gaining support within the open source community already and have contributors and support in not only the government and academic space but also the commercial sector (e.g. Intel). Being both open source and vendor neutral puts Warewulf in a unique position to be supported and utilized as an open standard in High Performance Computing (HPC). Warewulf is an active member of various HPC initiatives and groups and is considered one of the most utilized, extensible and scalable clustering toolkits available.

Kurtzer, GregoryM.↗
DOE OSTI2017

Template Interfaces for Agile Parallel Data-Intensive Science

Tigres provides a programming library to compose and execute large-scale data-intensive scientific workflows from desktops to supercomputers. DOE User Facilities and large science collaborations are increasingly generating large enough data sets that it is no longer practical to download them to a desktop to operate on them. They are instead stored at centralized compute and storage resources such as high performance computing (HPC) centers. Analysis of this data requires an ability to run on these facilities, but with current technologies, scaling an analysis to an HPC center and to a large data set is difficult even for experts. Tigres is addressing the challenge of enabling collaborative analysis of DOE Science data through a new concept of reusable "templates" that enable scientists to easily compose, run and manage collaborative computational tasks. These templates define common computation patterns used in analyzing a data set.

Ramakrishnan, Lavanya↗