Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “runtime performance”

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 235 records · Page 13
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Integrating and Characterizing HPC Task Runtime Systems for hybrid AI-HPC workloads

Scientific workflows increasingly involve both HPC and machine-learning tasks, combining MPI-based simulations, training, and inference in a single execution. Launchers such as Slurm’s srun constrain concurrency and throughput, making them unsuitable for dynamic and heterogeneous workloads. We present a performance study of RADICAL-Pilot (RP) integrated with Flux and Dragon, two complementary runtime systems that enable hierarchical resource management and high-throughput function execution. Using synthetic and production-scale workloads on Frontier, we characterize the task execution properties of RP across runtime configurations. RP+Flux sustains up to 930 tasks/s, and RP+Flux+Dragon exceeds 1,500 tasks/s with over 99.6% utilization. In contrast, srun peaks at 152 tasks/s and degrades with scale, with utilization below 50%. For IMPECCABLE.v2 drug discovery campaign, RP+Flux reduces makespan by 30–60% relative to srun/Slurm and increases throughput more than four times on up to 1,024. These results demonstrate hybrid runtime integration in RP as a scalable approach for hybrid AI-HPC workloads.

HPC-AI↗
DOE OSTI2024

IRIS Reimagined: Advancements in Intelligent Runtime System for Task-Based Programming

Task-based programming models are gaining traction in scientific computing. IRIS is a portable runtime system that exploits multiple heterogeneous programming systems and can discover available resources and manage multiple diverse programming systems (e.g., CUDA, Hexagon, HIP, Level Zero, OpenCL, and OpenMP) simultaneously. It accounts for the constraints of task dependencies and provides customizable scheduling policies to map those tasks to heterogeneous devices. In this paper, we present new capabilities added to IRIS to improve its portability for heterogeneous programming, build-friendliness, and performance efficiency. The new additions include vendor-specific kernel support, a runtime system with a foreign function interface to eliminate writing wrapper or boilerplate code for heterogeneous kernels, an easy-to-use and configurable CMake-based build environment, automatic and efficient data transfers and orchestration, and the Hunter and DAGGER toolchains to evaluate IRIS’s task scheduling algorithms.

Miniskar, Narasinga Rao↗
DOE OSTI2018

Automatic Halo Management for the Uintah GPU-Heterogeneous Asynchronous Many-Task Runtime

The Uintah computational framework is used for the parallel solution of partial differential equations on adaptive mesh refinement grids using modern supercomputers. Uintah is structured with an application layer and a separate runtime system. Uintah is based on a distributed directed acyclic graph of computational tasks, with a task scheduler that efficiently schedules and executes these tasks on both CPU cores and on-node accelerators. The runtime system identifies task dependencies, creates a task graph prior to the execution of these tasks, automatically generates MPI message tags, and automatically performs halo transfers for simulation variables. Automating halo transfers in a heterogeneous environment poses significant challenges when tasks compute within a few milliseconds, as runtime overhead affects wall time execution, or when simulation variables require large halos spanning most or all of the computational domain, as task dependencies become expensive to process. These challenges are magnified at production scale when application developers require each compute node perform thousands of different halo transfers among thousands simulation variables. The principal contribution of this work is to (1) identify and address inefficiencies that arise when mapping tasks onto the GPU in the presence of automated halo transfers, (2) implement new schemes to reduce runtime system overhead, (3) minimize application developer involvement with the runtime, and (4) show overhead reduction results from these improvements.

Concurrency↗
DOE OSTI2017

GASNet Specification, v1.8.1

GASNet is a language-independent, low-level networking layer that provides network-independent, high-performance communication primitives tailored for implementing parallel global address space SPMD languages and libraries such as UPC, UPC++, Co-Array Fortran, Legion, Chapel, and many others. The interface is primarily intended as a compilation target and for use by runtime library writers (as opposed to end users), and the primary goals are high performance, interface portability, and expressiveness. GASNet stands for "Global-Address Space Networking".

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

DYFLOW: A flexible framework for orchestrating scientific workflows on supercomputers

Modern scientific workflows are increasing in complexity with growth in computation power, incorporation of non-traditional computation methods, and advances in technologies enabling data streaming to support on-the-fly computation. These workflows have unpredictable runtime behaviors, and a fixed, predetermined resource assignment on supercomputers can be inefficient for overall performance and throughput. Inability to change resource assignments further limits the scientists to avail of science-driven opportunities or respond to failures.We introduce DYFLOW, a flexible framework that orchestrates scientific workflows on supercomputers based on user-designed policies. DYFLOW compartmentalizes orchestration stages into simplified constructs, and end-users can program and reuse them according to their workflow requirements through an easy-to-use interface. These constructs hide the intricacies involved in runtime management from end-users, for instance, procurement of information to understand the workflow state, assessment, and supervision of the runtime changes. DYFLOW is designed to work alongside existing workflow management systems and reuse the available (static) support for workflow management. We have integrated DYFLOW with an existing workflow management tool as a demonstration. With experiments performed on use cases from three types of scientific workflows and two different parallel architectures, we show that DYFLOW achieves the desired orchestration incurring a small cost to carry out the runtime changes.

Singhal, Swati↗
DOE OSTI2024

An open-source framework for balancing computational speed and fidelity in production cost models

Studies of bulk power system operations need to incorporate uncertainty and sensitivity analyses, especially around exposure to weather and climate variability and extremes, but this remains a computational modeling challenge. Commercial production cost models (PCMs) have shorter runtimes, but also important limitations (opacity, license restrictions) that do not fully support stochastic simulation. Open-source PCMs represent a potential solution. They allow for multiple, simultaneous runs in high-performance computing environments and offer flexibility in model parameterization. Yet, developers must balance computational speed (i.e. runtime) with model fidelity (i.e. accuracy). In this paper, we present Grid Operations (GO), a framework for instantiating open-source, scale-adaptive PCMs. GO allows users to search across parameter spaces to identify model versions that appropriately balance computational speed and fidelity based on experimental needs and resource limits. Results provide generalizable insights on how to navigate the fidelity and computational speed tradeoff through parameter selection. We show that models with coarser network topologies can accurately mimic market operations, sometimes better than higher-resolution models. It is thus possible to conduct large simulation experiments that characterize operational risks related to climate and weather extremes while maintaining sufficient model accuracy.

42 ENGINEERING↗
DOE OSTI2023

Evaluation of native Earth system model output with ESMValTool v2.6.0

Earth system models (ESMs) are state-of-the-art climate models that allow numerical simulations of the past, present-day, and future climate. To extend our understanding of the Earth system and improve climate change projections, the complexity of ESMs heavily increased over the last decades. As a consequence, the amount and volume of data provided by ESMs has increased considerably. Innovative tools for a comprehensive model evaluation and analysis are required to assess the performance of these increasingly complex ESMs against observations or reanalyses. One of these tools is the Earth System Model Evaluation Tool (ESMValTool), a community diagnostic and performance metrics tool for the evaluation of ESMs. Input data for ESMValTool needs to be formatted according to the CMOR (Climate Model Output Rewriter) standard, a process that is usually referred to as “CMORization”. While this is a quasi-standard for large model intercomparison projects like the Coupled Model Intercomparison Project (CMIP), this complicates the application of ESMValTool to non-CMOR-compliant climate model output. In this paper, we describe an extension of ESMValTool introduced in v2.6.0 that allows seamless reading and processing of “native” climate model output, i.e., operational output produced by running the climate model through the standard workflow of the corresponding modeling institute. This is achieved by an extension of ESMValTool's preprocessing pipeline that performs a CMOR-like reformatting of the native model output during runtime. Thus, the rich collection of diagnostics provided by ESMValTool is now fully available for these models. For models that use unstructured grids, a further preprocessing step required to apply many common diagnostics is regridding to a regular latitude–longitude grid. Extensions to ESMValTool's regridding functions described here allow for more flexible interpolation schemes that can be used on unstructured grids. Currently, ESMValTool supports nearest-neighbor, bilinear, and first-order conservative regridding from unstructured grids to regular grids. Example applications of this new native model support are the evaluation of new model setups against predecessor versions, assessing of the performance of different simulations against observations, CMORization of native model data for contributions to model intercomparison projects, and monitoring of running climate model simulations. For the latter, new general-purpose diagnostics have been added to ESMValTool that are able to plot a wide range of variable types. Currently, five climate models are supported: CESM2 (experimental; at the moment, only surface variables are available), EC-Earth3, EMAC, ICON, and IPSL-CM6. As the framework for the CMOR-like reformatting of native model output described here is implemented in a general way, support for other climate models can be easily added.

58 GEOSCIENCES↗
DOE OSTI2017

UNITY: Unified Memory and File Space. In: ROSS '17 Proceedings of the 7th International Workshop on Runtime and Operating Systems for Supercomputers ROSS 2017, Article No. 6

This paper describes the vision for UNITY, a new high-performance computing focused data storage abstraction that places the entire memory hierarchy, including both traditionally separated memory-and file-based data storage, into one storage continuum. Through the use of a novel API and a set of services centered around a smart runtime system, UNITY is able to provide a number of valuable and interesting benefits. The unified storage space provides a scalable and resilient data environment that dynamically manages the mapping of data onto available resources based on multiple factors, including desired persistence and energy budget considerations. By eliminating the need for high-performance computing domain scientists to develop architecture-dependent optimizations for rapidly evolving data storage technologies, UNITY addresses both ease-of-use and performance.

Jones, Terry↗
DOE OSTI2018

Global-Address Networking (GASNet-1) Library

GASNet is a language-independent, networking middleware layer that provides network-independent, high-performance communication primitives including Remote Memory Access (RMA) and Active Messages (AM). It has been used to implement parallel programming models and libraries such as UPC, Co-Array Fortran, Titanium, Legion, Chapel, and many others. The interface is primarily intended as a compilation target and for use by runtime library writers (as opposed to end users), and the primary goals are high performance, interface portability, and expressiveness. GASNet stands for "Global-Address Space Networking".

SciDAC↗
DOE OSTI2025

Predicting runtime and resource utilization of jobs on integrated cloud and HPC systems

Recent advances in virtualization technologies used in cloud computing offer performance that closely approaches bare-metal levels. Combined with specialized instance types and high-speed networking services for cluster computing, cloud platforms have become a compelling option for high-performance computing (HPC). However, most current batch job schedulers in HPC systems are designed for homogeneous clusters and make decisions based on limited information about jobs and system status. Scientists typically submit computational jobs to these schedulers with a requested runtime that is often over- or under-estimated. More accurate runtime predictions can help schedulers make better decisions and reduce job turnaround times. Here, they can also support decisions about migrating jobs to the cloud to avoid long queue wait times in HPC systems.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2015

GPU-based simulation of the two-dimensional unstable structure of gaseous oblique detonations

In this paper, the two-dimensional structure of unstable oblique detonations induced by the wedge from a supersonic combustible gas flow is simulated using the reactive Euler equations with a one-step Arrhenius chemistry model. A wide range of activation energy of the combustible mixture is considered. Computations are performed on the Graphical Processing Unit (GPU) to reduce the simulation runtimes. A large computational domain covered by a uniform mesh with high grid resolution is used to properly capture the development of instabilities and the formation of different transverse wave structures. After the initiation point, where the oblique shock transits into a detonation, an instability begins to manifest and in all cases, the left-running transverse waves first appear, followed by the subsequent emergence of right-running transverse waves forming the dual-head triple point structure. This study shows that for low activation energies, a long computational length must be carefully considered to reveal the unstable surface due to the slow growth rate of the instability. For high activation energies, the flow behind the unstable oblique detonation features the formation of unburnt gas pockets and strong vortex-pressure wave interaction resulting in a chaotic-like vortical structure.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2018

Fast Determination of Distribution-Connected PV Impacts Using a Variable-Time-Step Quasi-Static Time-Series Approach

The increasing deployment of distribution-connected photovoltaic (DPV) systems requires utilities to complete ever more complex interconnection studies. Relatively simple interconnection study methods worked well for low penetrations of photovoltaic systems, but more complicated quasistatic time-series (QSTS) analysis is required to make better interconnection decisions as DPV penetration levels increase. Tools and methods must be developed to support this type of complex analysis to enable efficient completion of PV impact studies. This paper presents a variable-time-step solver for QSTS analysis that significantly shortens the computational time and effort to complete a detailed analysis of the operation of a distribution circuit with many DPV systems. Specifically, it demonstrates that the proposed variable-time-step solver can reduce the required computational time by as much as 84% without introducing any important errors to metrics, such as the number of voltage regulator tap operations and highest and lowest voltage occurring on the distribution circuit during a 1-yr simulation period. Further improvement in computational speed is possible with the introduction of only modest errors in these metrics, such as a 91% reduction with less than 5% error when predicting voltage regulator operations. Furthermore, an alternate variable time-step method is also shown to perform adequately and shows promise of even greater reductions in simulation runtime.

14 SOLAR ENERGY↗
DOE OSTI2025

SDA: a symbolic differential algebra package in C++

Truncated Power Series Algebra (TPSA), or Differential Algebra (DA), is a well-established tool in accelerator physics, commonly used for generating high-order maps of dynamic systems, as well as in symplectic tracking, normal form analysis, verified integration, optimization, and fast multipole methods. This package is the first to perform symbolic DA computations, enabling traceability of initial condition contributions and runtime reduction for repeated DA calculations, potentially expanding DA’s applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Enhancements to the Subplane Capability in MPACT

In recent years, much of the development effort for the Virtual Environment for Reactor Applications (VERA) has been devoted to improving performance and targeting industry-class computers. This includes reducing both runtime (improving efficiency) and core-count requirements to enable simulation on smaller machines. In MPACT, in particular, many aspects of the neutronic iteration strategy have been evaluated and improved, such as the subgroup and method of characteristics (MOC) runtimes, and many efforts have focused on improving the coarse mesh finite difference (CMFD) runtime, which has become a dominant component of the runtime as other aspects have been improved.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

A Brief Survey on High Performance Computing Systems Power Management

This paper provides a survey of software-based power management techniques in High Performance Computing (HPC) systems. Seven existing power management and monitoring tools and frameworks are discussed. These are: Variorum, dynamic energy-performance optimizer (DEPO), Powersched, Bull Dynamic Power Optimizer (BDPO), Energy Aware Runtime (EAR), Global Extensible Open Power Manager (GEOPM), and PoLiMEr. Each of these tools is evaluated based on hardware abstraction, optimization methods, usability, and experimental validation. This survey highlights the diversity of approaches in managing energy efficiency, from vendor-neutral APIs to algorithm-driven power capping, and dynamic frequency adjustments. Given that energy requirements for large computational systems is increasing quickly, the importance of integrating these tools into existing HPC environments and the need for further research in this rapidly evolving field is also discussed.

97 - MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Characterizing the Performance of Executing Many-tasks on Summit

Many scientific workloads are comprised of many tasks, where each task is an independent simulation or analysis of data. The execution of millions of tasks on heterogeneous HPC platforms requires scalable dynamic resource management and multi-level scheduling. RADICAL-Pilot (RP) -- an implementation of the Pilot abstraction, addresses these challenges and serves as an effective runtime system to execute workloads comprised of many tasks. In this paper, we characterize the performance of executing many tasks using RP when interfaced with JSM and PRRTE on Summit: RP is responsible for resource management and task scheduling on acquired resource; JSM or PRRTE enact the placement of launching of scheduled tasks. Our experiments provide lower bounds on the performance of RP when integrated with JSM and PRRTE. Specifically, for workloads comprised of homogeneous single-core, 15 minutes-long tasks we find that: PRRTE scales better than JSM for > O(1000) tasks; PRRTE overheads are negligible; and PRRTE supports optimizations that lower the impact of overheads and enable resource utilization of 63% when executing O(16K), 1-core tasks over 404 compute nodes.

Matteo, Turilli↗
DOE OSTI2015

On the Impact of Execution Models: A Case Study in Computational Chemistry

Efficient utilization of high-performance computing (HPC) platforms is an important and complex problem. Execution models, abstract descriptions of the dynamic runtime behavior of the execution stack, have significant impact on the utilization of HPC systems. Using a computational chemistry kernel as a case study and a wide variety of execution models combined with load balancing techniques, we explore the impact of execution models on the utilization of an HPC system. We demonstrate a 50 percent improvement in performance by using work stealing relative to a more traditional static scheduling approach. We also use a novel semi-matching technique for load balancing that has comparable performance to a traditional hypergraph-based partitioning implementation, which is computationally expensive. Using this study, we found that execution model design choices and assumptions can limit critical optimizations such as global, dynamic load balancing and finding the correct balance between available work units and different system and runtime overheads. With the emergence of multi- and many-core architectures and the consequent growth in the complexity of HPC platforms, we believe that these lessons will be beneficial to researchers tuning diverse applications on modern HPC platforms, especially on emerging dynamic platforms with energy-induced performance variability.

Execution models, computational chemistry, MPI and↗