Search NASA⌕ Search

SEARCH · Search NASA

Results for “runtime systems”

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 739 records · Page 41
All sourcesNASA NTRSDOE OSTI
NASA NTRS2013

Maintaining the Health of Software Monitors

Software health management (SWHM) techniques complement the rigorous verification and validation processes that are applied to safety-critical systems prior to their deployment. These techniques are used to monitor deployed software in its execution environment, serving as the last line of defense against the effects of a critical fault. SWHM monitors use information from the specification and implementation of the monitored software to detect violations, predict possible failures, and help the system recover from faults. Changes to the monitored software, such as adding new functionality or fixing defects, therefore, have the potential to impact the correctness of both the monitored software and the SWHM monitor. In this work, we describe how the results of a software change impact analysis technique, Directed Incremental Symbolic Execution (DiSE), can be applied to monitored software to identify the potential impact of the changes on the SWHM monitor software. The results of DiSE can then be used by other analysis techniques, e.g., testing, debugging, to help preserve and improve the integrity of the SWHM monitor as the monitored software evolves.

Runtime Monitor↗
NASA NTRS1993

Efficiently modeling neural networks on massively parallel computers

Neural networks are a very useful tool for analyzing and modeling complex real world systems. Applying neural network simulations to real world problems generally involves large amounts of data and massive amounts of computation. To efficiently handle the computational requirements of large problems, we have implemented at Los Alamos a highly efficient neural network compiler for serial computers, vector computers, vector parallel computers, and fine grain SIMD computers such as the CM-2 connection machine. This paper describes the mapping used by the compiler to implement feed-forward backpropagation neural networks for a SIMD (Single Instruction Multiple Data) architecture parallel computer. Thinking Machines Corporation has benchmarked our code at 1.3 billion interconnects per second (approximately 3 gigaflops) on a 64,000 processor CM-2 connection machine (Singer 1990). This mapping is applicable to other SIMD computers and can be implemented on MIMD computers such as the CM-5 connection machine. Our mapping has virtually no communications overhead with the exception of the communications required for a global summation across the processors (which has a sub-linear runtime growth on the order of O(log(number of processors)). We can efficiently model very large neural networks which have many neurons and interconnects and our mapping can extend to arbitrarily large networks (within memory limitations) by merging the memory space of separate processors with fast adjacent processor interprocessor communications. This paper will consider the simulation of only feed forward neural network although this method is extendable to recurrent networks.

Farber, Robert M.↗
NASA NTRS2002

Reducing False Positives in Runtime Analysis of Deadlocks

This paper presents an improvement of a standard algorithm for detecting dead-lock potentials in multi-threaded programs, in that it reduces the number of false positives. The standard algorithm works as follows. The multi-threaded program under observation is executed, while lock and unlock events are observed. A graph of locks is built, with edges between locks symbolizing locking orders. Any cycle in the graph signifies a potential for a deadlock. The typical standard example is the group of dining philosophers sharing forks. The algorithm is interesting because it can catch deadlock potentials even though no deadlocks occur in the examined trace, and at the same time it scales very well in contrast t o more formal approaches to deadlock detection. The algorithm, however, can yield false positives (as well as false negatives). The extension of the algorithm described in this paper reduces the amount of false positives for three particular cases: when a gate lock protects a cycle, when a single thread introduces a cycle, and when the code segments in different threads that cause the cycle can actually not execute in parallel. The paper formalizes a theory for dynamic deadlock detection and compares it to model checking and static analysis techniques. It furthermore describes an implementation for analyzing Java programs and its application to two case studies: a planetary rover and a space craft altitude control system.

Bensalem, Saddek↗
DOE OSTI2020

DPM: A deep learning PDE augmentation method with application to large-eddy simulation

A framework is introduced that leverages known physics to reduce overfitting in machine learning for scientific applications. The partial differential equation (PDE) that expresses the physics is augmented with a neural network that uses available data to learn a description of the corresponding unknown or unrepresented physics. Training within this combined system corrects for missing, unknown, or erroneously represented physics, including discretization errors associated with the PDE's numerical solution. For optimization of the network within the PDE, an adjoint PDE is solved to provide high-dimensional gradients, and a stochastic adjoint method (SAM) further accelerates training. Additionally, the approach is demonstrated for large-eddy simulation (LES) of turbulence. High-fidelity direct numerical simulations (DNS) of decaying isotropic turbulence provide the training data used to learn sub-filter-scale closures for the filtered Navier–Stokes equations. Out-of-sample comparisons show that the deep learning PDE method outperforms widely-used models, even for filter sizes so large that they become qualitatively incorrect. It also significantly outperforms the same neural network when a priori trained based on simple data mismatch, not accounting for the full PDE. Measures of discretization errors, which are well-known to be consequential in LES, point to the importance of the unified training formulation's design, which without modification corrects for them. For comparable accuracy, simulation runtime is significantly reduced. A relaxation of the typical discrete enforcement of the divergence-free constraint in the solver is also successful, instead allowing the DPM to approximately enforce incompressibility physics. Since the training loss function is not restricted to correspond directly to the closure to be learned, training can incorporate diverse data, including experimental data.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2019

Presentation layer of CMS Online Monitoring System

The Compact Muon Solenoid (CMS) is one of the experiments at the CERN Large Hadron Collider (LHC). The CMS Online Monitoring system (OMS) is an upgrade and successor to the CMS Web-Based Monitoring (WBM)system, which is an essential tool for shift crew members, detector subsystem experts, operations coordinators, and those performing physics analyses. The CMS OMS is divided into aggregation and presentation layers. Communication between layers uses RESTful JSON:API compliant requests. The aggregation layer is responsible for collecting data from heterogeneous sources, storage of transformed and pre-calculated (aggregated) values and exposure of data via the RESTful API. The presentation layer displays detector information via a modern, user-friendly and customizable web interface. The CMS OMS user interface is composed of a set of cutting-edge software frameworks and tools to display non-event data to any authenticated CMS user worldwide. The web interface tree-like component structure comprises (top-down): workspaces, folders, pages, controllers and portlets. A clear hierarchy gives the required flexibility and control for content organization. Each bottom element instantiates a portlet and is a reusable component that displays a single aspect of data, like a table, a plot, an article, etc. Pages consist of multiple different portlets and can be customized at runtime by using a drag-and-drop technique. This is how a single page can easily include information from multiple online sources. Different pages give access to a summary of the current status of the experiment, as well as convenient access to historical data. This paper describes the CMS OMS architecture, core concepts and technologies of the presentation layer.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2026

Reduced-order CFD modeling of cryogenic hydrogen isotope extrusion for pellet fueling

This study presents a reduced-order model (ROM) for computational fluid dynamics (CFD) simulations of cryogenic hydrogen isotope extrusions, focusing on protium (H₂) and deuterium (D₂) piston extruders. Using a 2D axisymmetric ROM in ANSYS-Polyflow, significant computational savings were achieved (runtime reduced from 9∼24 h to 3∼5 min), with extrusion force discrepancies between the 2D ROM and 3D models being on the order of 1%. Parametric studies identified optimal cutoff shear rates in the viscosity model (0.01/s for H₂ and 0.001/s for D₂), providing recommendations for future simulations. Finally, a comprehensive comparison of ROM results with experimental data was performed across varying geometries, cryogenic materials, temperatures, extrusion lengths, and piston velocities. Predictions at low extrusion temperatures met the objective of providing quick and efficient solutions with an acceptable extrusion force error of approximately 10% or less, validating the effectiveness of the 2D ROM approach. However, at high temperatures closer to the triple point, extrusion force error grows, which necessitates developing an improved model that accounts for temperature effects, e.g. melting. Nevertheless, the findings still represent a significant improvement in efficiency of CFD modeling of cryogenic hydrogenic extrusion. The ROM framework can also be extended to tritium (T2) and screw extruders, which will ultimately provide a fast and effective tool for optimizing pellet injector design for ITER and future reactor systems.

Fan, Joy [ORNL] (ORCID:0000000229751735)↗
DOE OSTI2022

TIA: A forward model and analyzer for Talbot interferometry experiments of dense plasmas

Interferometry is one of the most sensitive and successful diagnostic methods for plasmas. However, owing to the design of most common interferometric systems, the wavelengths of operation and, therefore, the range of densities and temperatures that can be probed are severely limited. Talbot–Lau interferometry offers the possibility of extending interferometry measurements to x-ray wavelengths by means of the Talbot effect. While there have been several proof-of-concept experiments showing the efficacy of this method, it is only recently that experiments to probe High Energy Density (HED) plasmas using Talbot–Lau interferometry are starting to take place. To improve these experimental designs, we present here the Talbot-Interferometry Analyzer (TIA) tool, a forward model for generating and postprocessing synthetic x-ray interferometry images from a Talbot–Lau interferometer. Although TIA can work with any two-dimensional hydrodynamic code to study plasma conditions as close to reality as possible, this software has been designed to work by default with output files from the hydrodynamic code FLASH, making the tool user-friendly and accessible to the general plasma physics community. Here, the model has been built into a standalone app, which can be installed by anyone with access to the MATLAB runtime installer and is available upon request to the authors.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
NASA NTRSDate not supplied

Rapid Spacecraft Payload Development: In-Orbit Demonstration of Flight Software Reuse, Scalability, and Dependability

As space mission design trends towards shared, multi-mission platforms and high-performance onboard computing architectures, the number of spacecraft launched into operation is also steadily rising. Through ridesharing, spacecraft miniaturization, and other cost-reduction measures, the barriers to space are lowering, resulting in compounded growth in the amount of flight software being deployed. To meet the needs of both the growing quantity and evolving nature of spacecraft, flight software design must accordingly adapt to support more efficient development, solutions to computational resource-sharing, and software reusability. This paper focuses on a software payload demonstrating several core technologies that improve the state-of-the-art in these identified areas. Launched into low-earth orbit in January 2022, our software payload was conceived, designed, and delivered in a span of merely two months. It was developed on top of the NASA core Flight System (cFS) framework and the Distributed Spacecraft Autonomy (DSA) Comm cFS application, which translates cFS software bus messages across a Data Distribution Service (DDS) network. The flight software, packaged in Linux container images, was deployed as one of 18 flight applications managed through the Unibap SpaceCloud Framework. The applications were run on a Unibap iX5-102 radiation-tolerant payload computer, hosted on the D-Orbit SCV-004 spacecraft as part of an ESA-sponsored in-orbit technology test. Our payload, referred to as the DSA D-Orbit software, demonstrates the reusability of the DSA Comm app in a substantially different context and purpose as its original mission. Comm’s original design goal was to reliably distribute messages between spacecraft swarms of arbitrary size and dynamic network topology. However, we leverage this same functionality to introduce redundancy and opportunistic parallel data processing in the context of a representative onboard image processing workload. This adaptive mission architecture was enabled in part by the SpaceCloud Framework’s use of container virtualization as the payload integration interface. By using a base container image with common high-level language runtimes and libraries, we were able to rapidly design, develop, and validate our image processing application without many of the technological barriers common to flight software development. We present details the goals, approach, results, and lessons learned through this technology demonstration experiment and contextualize those observations against present and future challenges in spacecraft software development.

computer programming↗
NASA NTRS2018

New Runtime Record Set for Free-Piston Stirling Machine

This article describes the recent progress GRC has made in the arena of long-life heat engines for power conversion. NASA has been interested in Stirling-cycle machines for power conversion since the 1970s. At that time, it was recognized that such a device could be designed to achieve long-life continuous operation with high reliability, by leveraging the free-piston configuration in favor of a kinematic engine. The free-piston variant eliminates the wear mechanisms associated with rolling-element bearings, piston rings, and connecting rods, typically found in kinematic Stirling engine designs. By restricting the moving components to linear motion only, there is no need to convert linear piston movement to rotating motion. With this, the moving components can be suspended within their cylinders by non-contacting bearings and seals can be effected with close clearances, also non-contacting. The use of a linear alternator attached directly to the piston enables direct conversion of heat, to mechanical, then to electrical energy, all in one hermetically sealed vessel, The electrical output only requires simple feedthroughs, and the complexity of sealed a rotating crankshaft is eliminated. In the free-piston variant, the displacer-to-piston phase is not implemented by a connection to a common rotating shaft. Instead, the pressure wave itself is used to drive the displacer. The displacer is designed to have a higher natural frequency of motion than the piston, which achieves a leading phase angle required for power production. The free-piston Stirling behaves like a spring-mass-damper system, with a forcing function created by the alternate heater and cooling of the working gas, and damping provided by the current flowing through the linear alternator. Two types of non-contacting bearings are available to designers: flexure and gas bearings. Flexure bearings consist of a spiral-cut disc of metal that has high radial stiffness but a deliberately engineering stiffness in the axial direction, to permit piston and displacer resonant motion. Gas bearings utilize a pressure source and precisely machined flow channels to provide a radial stiffness around the moving component. The pressure source can be supplied by the engine's pressure wave itself.

Oriti, Salvatore M.↗
DOE OSTI2026

A Survey on the Expanding Scope and Interdisciplinary Opportunities for Processing-in-Memory Techniques

Processing-in-Memory (PIM) is emerging as a practical path to overcome the limitations of traditional von Neumann architectures. At its core, PIM systems implement computing primitives such as logic operations and multiply-accumulate acceleration through compute-in-memory, near-memory processing, or hybrid designs. The role of memory cells varies widely across technologies, acting as inputs, outputs, or analog accumulators through bit-lines and sense amplifiers. This diversity creates trade-offs in precision, bandwidth, latency, and programmability, making it difficult to build a unified understanding on the progress of the field. In this survey, we organize recent advances of PIM into three areas. First, we discuss the progress on the architectural optimizations of PIM and its integration with both DRAM and emerging non-volatile memories. Second, we examine how PIM is being used to accelerate key computing domains, including generative AI workloads and high-performance kernels, along with new approaches. Third, we highlight the growing adoption of PIM in computational sciences, where it is being applied to solve interdisciplinary problems such as genome analysis, mRNA quantification, mass spectrometry, quantum circuit simulation, wave modeling, and secure computation. Finally, we synthesize the major challenges that continue to slow PIM adoption, including manufacturing constraints, power delivery, thermal reliability, data consistency, runtime and memory-management coordination, and the difficulty of building portable software abstractions without sacrificing commercial viability. This work provides an updated, structured perspective on PIM’s potential across computing and computational sciences and the barriers that must be solved for it to reach its full impact.

Asifuzzaman, Kazi [Oak Ridge National Laboratory (↗
DOE OSTI2025

Efficient and assured reinforcement learning-based building HVAC control with heterogeneous expert-guided training

Abstract Building heating, ventilation, and air conditioning (HVAC) systems account for nearly half of building energy consumption and $$20\%$$ of total energy consumption in the US. Their operation is also crucial for ensuring the physical and mental health of building occupants. Compared with traditional model-based HVAC control methods, the recent model-free deep reinforcement learning (DRL) based methods have shown good performance while do not require the development of detailed and costly physical models. However, these model-free DRL approaches often suffer from long training time to reach a good performance, which is a major obstacle for their practical deployment. In this work, we present a systematic approach to accelerate online reinforcement learning for HVAC control by taking full advantage of the knowledge from domain experts in various forms . Specifically, the algorithm stages include learning expert functions from existing abstract physical models and from historical data via offline reinforcement learning, integrating the expert functions with rule-based guidelines, conducting training guided by the integrated expert function and performing policy initialization from distilled expert function. Moreover, to ensure that the learned DRL-based HVAC controller can effectively keep room temperature within the comfortable range for occupants, we design a runtime shielding framework to reduce the temperature violation rate and incorporate the learned controller into it. Experimental results demonstrate up to 8.8 X speedup in DRL training from our approach over previous methods, with low temperature violation rate.

Xu, Shichao↗
NASA NTRS1997

Performance Data Gathering and Representation from Fixed-Size Statistical Data

The two commonly-used performance data types in the super-computing community, statistics and event traces, are discussed and compared. Statistical data are much more compact but lack the probative power event traces offer. Event traces, on the other hand, are unbounded and can easily fill up the entire file system during program execution. In this paper, we propose an innovative methodology for performance data gathering and representation that offers a middle ground. Two basic ideas are employed: the use of averages to replace recording data for each instance and 'formulae' to represent sequences associated with communication and control flow. The user can trade off tracing overhead, trace data size with data quality incrementally. In other words, the user will be able to limit the amount of trace data collected and, at the same time, carry out some of the analysis event traces offer using space-time views. With the help of a few simple examples, we illustrate the use of these techniques in performance tuning and compare the quality of the traces we collected with event traces. We found that the trace files thus obtained are, indeed, small, bounded and predictable before program execution, and that the quality of the space-time views generated from these statistical data are excellent. Furthermore, experimental results showed that the formulae proposed were able to capture all the sequences associated with 11 of the 15 applications tested. The performance of the formulae can be incrementally improved by allocating more memory at runtime to learn longer sequences.

Yan, Jerry C.↗
DOE OSTI2022

Monitoring of Temperature Measurements for Different Flow Regimes in Water and Galinstan with Long Short-Term Memory Networks and Transfer Learning of Sensors

Temperature sensing is one of the most common measurements of a nuclear reactor monitoring system. The coolant fluid flow in a reactor core depends on the reactor power state. We investigated the monitoring and estimation of the thermocouple time series using machine learning for a range of flow regimes. Measurement data were obtained, in two separate experiments, in a flow loop filled with water and with liquid metal Galinstan. We developed long short-term memory (LSTM) recurrent neural networks (RNNs) for sensor predictions by training on the sensor’s own prior history, and transfer learning LSTM (TL-LSTM) by training on a correlated sensor’s prior history. Sensor cross-correlations were identified by calculating the Pearson correlation coefficient of the time series. The accuracy of LSTM and TL-LSTM predictions of temperature was studied as a function of Reynolds number (Re). The root-mean-square error (RMSE) for the test segment of time series of each sensor was shown to linearly increase with Re for both water and Galinstan fluids. Using linear correlations, we estimated the range of values of Re for which RMSE is smaller than the thermocouple measurement uncertainty. For both water and Galinstan fluids, we showed that both LSTM and TL-LSTM provide reliable estimations of temperature for typical flow regimes in a nuclear reactor. The LSTM runtime was shown to be substantially smaller than the data acquisition rate, which allows for performing estimation and validation of sensor measurements in real time.

Pantopoulou, Stella↗
DOE OSTI2023

Toward Consistent High-Fidelity Quantum Learning on Unstable Devices via Efficient In-Situ Calibration

In the near-term noisy intermediate-scale quantum (NISQ) era, high noise will significantly reduce the fidelity of quantum computing. What's worse, recent works reveal that the noise on quantum devices is not stable, that is, the noise is dynamically changing over time. This leads to an imminent challenging problem: At run-time, is there a way to efficiently achieve a consistent high-fidelity quantum system on unstable devices? To study this problem, we take quantum learning (a.k.a., variational quantum algorithm) as a vehicle, which has a wide range of applications, such as combinatorial optimization and machine learning. A straightforward approach is to optimize a variational quantum circuit (VQC) with a parameter-shift approach on the target quantum device before using it; however, the optimization has an extremely high time cost, which is not practical at run-time. To address the pressing issue, in this paper, we proposed a novel quantum pulse-based noise adaptation framework, namely QuPAD. In the proposed framework, first, we identify that the CNOT gate is the fidelity bottleneck of the conventional VQC, and we employ a more robust parameterized multi-qubit gate (i.e., Rzx gate) to replace CNOT gate. Second, by benchmarking Rzx gate with different parameters, we build a fitting function for each coupling qubit pair, such that the deviation between the theoretic output of Rzx gate and its on-device output under a given pulse amplitude and duration can be efficiently predicted. On top of this, an evolutionary algorithm is devised to identify the pulse amplitude and duration of each Rzx gate (i.e., calibration) and find the quantum circuits with high fidelity. Experiments show that the runtime on quantum devices of QuPAD with 8–10 qubits is less than 15 minutes, which is up to 270 x faster than the parameter-shift approach. In addition, compared to the vanilla VQC as a baseline, QuPAD can achieve 59.33% accuracy gain on a classification task, and average 66.34% closer to ground state energy for molecular simulation.

Hu, Zhirui↗
DOE OSTI2021

Field Validation of a Smart Energy Recovery Ventilation System Using Low-Cost Indoor Air Quality Sensors

This project is a field validation, using low-cost indoor air quality (IAQ) sensors, of a smart ventilation system that can help low-load homes in humid environments maintain acceptable indoor humidity conditions while providing adequate ventilation according to ASHRAE 62.2. The objectives of this research were to (1) address builders’ concerns with mechanical ventilation in humid environments and (2) answer the question of whether smart control logic helps with occupant comfort and the creation of a more acceptable indoor environment. To address the objectives of the study, the Southface team collected field data for one year in four Charleston, South Carolina, new construction homes in order to determine the differences in occupant comfort; comfort metrics; IAQ; and heating, ventilating, and air-conditioning (HVAC) energy consumption when toggling biweekly between an energy recovery ventilator (ERV) operating continuously and an ERV operating with smart, time-varying humidity control logic. The smart ventilation algorithm under consideration in this field test did create a less humid indoor environment on an annual basis as quantitatively measured through temperature and relative humidity (T/RH) readings, expressed most discernably as “percentage of time above 60% RH” and “percentage of time above 55°F dewpoint.” However, the difference it made was inconsistent during the spring, summer, and fall months, and it was only directionally consistent during the winter months. We suspect that this is primarily due to the long runtimes and concomitant dehumidification activity of the air-conditioning (A/C) units in response to the high sensible loads in Charleston. The effect of the smart ventilation algorithm was not discernable to the occupants in this study, as recorded through seasonal surveys.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Multi-Entity Simulation with CoSim Toolbox

Co-simulation is an analysis technique for linking multiple software models during runtime by facilitating data exchange and simulation time synchronization. There are numerous challenges when constructing an effective co-simulation including simulation tool installation, data management, and writing new models in a manner compatible with the co-simulation framework of choice. CoSim Toolbox is an integration of multiple pieces of software designed to make assembling such a co-simulation in HELICS easier. This report summarizes the existing capabilities of CoSim Toolbox and outlines future development plans.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Context-aware learning of hierarchies of low-fidelity models for multi-fidelity uncertainty quantification

Multi-fidelity Monte Carlo methods leverage low-fidelity and surrogate models for variance reduction to make tractable uncertainty quantification even when numerically simulating the physical systems of interest with high-fidelity models is computationally expensive. This work proposes a context-aware multi-fidelity Monte Carlo method that optimally balances the costs of training low-fidelity models with the costs of Monte Carlo sampling. It generalizes the previously developed context-aware bi-fidelity Monte Carlo method to hierarchies of multiple models and to more general types of low-fidelity models. When training low-fidelity models, the proposed approach takes into account the context in which the learned low-fidelity models will be used, namely for variance reduction in Monte Carlo estimation, which allows it to find optimal trade-offs between training and sampling to minimize upper bounds of the mean-squared errors of the estimators for given computational budgets. This is in stark contrast to traditional surrogate modeling and model reduction techniques that construct low-fidelity models with the primary goal of approximating well the high-fidelity model outputs and typically ignore the context in which the learned models will be used in upstream tasks. Further, the proposed context-aware multi-fidelity Monte Carlo method applies to hierarchies of a wide range of types of low-fidelity models such as sparse-grid and deep-network models. Numerical experiments with the gyrokinetic simulation code Gene show speedups of up to two orders of magnitude compared to standard estimators when quantifying uncertainties in small-scale fluctuations in confined plasma in fusion reactors. This corresponds to a runtime reduction from 72 days to four hours on one node of the Lonestar6 supercomputer at the Texas Advanced Computing Center.

42 ENGINEERING↗
DOE OSTI2023

Comparing the Performance of Julia on CPUs versus GPUs and Julia-MPI versus Fortran-MPI: a case study with MPAS-Ocean (Version 7.1)

Abstract. Some programming languages are easy to develop at the cost of slow execution, while others are fast at runtime but much more difficult to write. Julia is a programming language that aims to be the best of both worlds – a development and production language at the same time. To test Julia's utility in scientific high-performance computing (HPC), we built an unstructured-mesh shallow water model in Julia and compared it against an established Fortran-MPI ocean model, the Model for Prediction Across Scales–Ocean (MPAS-Ocean), as well as a Python shallow water code. Three versions of the Julia shallow water code were created: for single-core CPU, graphics processing unit (GPU), and Message Passing Interface (MPI) CPU clusters. Comparing identical simulations revealed that our first version of the Julia model was 13 times faster than Python using NumPy, where both used an unthreaded single-core CPU. Further Julia optimizations, including static typing and removing implicit memory allocations, provided an additional 10–20× speed-up of the single-core CPU Julia model. The GPU-accelerated Julia code was almost identical in terms of performance to the MPI parallelized code on 64 processes, an unexpected result for such different architectures. Parallelized Julia-MPI performance was identical to Fortran-MPI MPAS-Ocean for low processor counts and ranges from 2× faster to 2× slower for higher processor counts. Our experience is that Julia development is fast and convenient for prototyping but that Julia requires further investment and expertise to be competitive with compiled codes. We provide advice on Julia code optimization for HPC systems.

54 ENVIRONMENTAL SCIENCES↗