Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory benchmark”

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 307 records · Page 17
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Implementing Arbitrary/Common Concurrent Writes of CRCW PRAM

The Parallel Random Access Machines (PRAM) abstraction is the simplest and most elegant algorithmic model for the design and analysis of parallel algorithms. It consists of different models categorized based on the underlying memory access mode used, the most powerful of which is the Concurrent Read Concurrent Write (CRCW) model. A PRAM algorithm describes a series of rounds, each of which consists of a collection of operations that can be executed concurrently within the same time step. However, the lack of support for concurrent memory accesses and the prevalence of asynchronous programming models led to the belief that implementing CRCW PRAM algorithms is unattainable and prompted many to avoid this model except for theoretical studies of optimal performance.In this work, we study the arbitrary and common concurrent writes in the CRCW PRAM model and explore implementation challenges on general-purpose systems. Moreover, we examine current practices for implementing common/arbitrary concurrent writes and propose a new efficient lightweight and thread-safe method to implement concurrent writes through leveraging atomic instructions. To demonstrate the efficacy of our method, we developed OpenMP kernels for classical CRCW PRAM algorithms and provide experimental results and comparisons based on run time performance measured over the x86 multicore architecture. Our results show a performance speedup compared to current practices up to 4.5x across all our benchmarks.

Ghanim, Fady↗
NASA NTRS2018

Parallelization of a Six Degree of Freedom Entry Vehicle Trajectory Simulation Using OpenMP and OpenACC

The art and science of writing parallelized software, using methods such as Open Multi-Processing (OpenMP) and Open Accelerators (OpenACC), is dominated by computer scientists. Engineers and non-computer scientists looking to apply these techniques to their project applications face a steep learning curve, especially when looking to adapt their original single threaded software to run multi-threaded on graphics processing units (GPUs). There are significant changes in mindset that must occur; such as how to manage memory, the organization of instructions, and the use of if statements (also known as branching). The purpose of this work is twofold: 1) to demonstrate the applicability of parallelized coding methodologies, OpenMP and OpenACC, to tasks outside of the typical large scale matrix mathematics; and 2) to discuss, from an engineer’s perspective, the lessons learned from parallelizing software using these computer science techniques. This work applies OpenMP, on both multi-core central processing units (CPUs) and Intel® Xeon Phi™ 7210, and OpenACC on GPUs. These parallelization techniques are used to tackle the simulation of thousands of entry vehicle trajectories through the integration of six degree of freedom (DoF) equations of motion (EoM). The forces and moments acting on the entry vehicle, and used by the EoM, are estimated using multiple models of varying levels of complexity. Several benchmark comparisons are made on the execution of six DoF trajectory simulation: single thread Intel® Xeon® E5-2670 CPU, multi-thread CPU using OpenMP, multi-thread Xeon Phi™ 7210 using OpenMP, and multi-thread NVIDIA® Tesla® K40 GPU using OpenACC. These benchmarks are run on the Pleiades Supercomputer Cluster at the National Aeronautics and Space Administration (NASA) Ames Research Center (ARC), and a Xeon Phi™ 7210 node at NASA Langley Research Center (LaRC).

Green, Justin S.↗
NASA NTRS1994

Inducer analysis/pump model development

Current design of high performance turbopumps for rocket engines requires effective and robust analytical tools to provide design information in a productive manner. The main goal of this study was to develop a robust and effective computational fluid dynamics (CFD) pump model for general turbopump design and analysis applications. A finite difference Navier-Stokes flow solver, FDNS, which includes an extended k-epsilon turbulence model and appropriate moving zonal interface boundary conditions, was developed to analyze turbulent flows in turbomachinery devices. In the present study, three key components of the turbopump, the inducer, impeller, and diffuser, were investigated by the proposed pump model, and the numerical results were benchmarked by the experimental data provided by Rocketdyne. For the numerical calculation of inducer flows with tip clearance, the turbulence model and grid spacing are very important. Meanwhile, the development of the cross-stream secondary flow, generated by curved blade passage and the flow through tip leakage, has a strong effect on the inducer flow. Hence, the prediction of the inducer performance critically depends on whether the numerical scheme of the pump model can simulate the secondary flow pattern accurately or not. The impeller and diffuser, however, are dominated by pressure-driven flows such that the effects of turbulence model and grid spacing (except near leading and trailing edges of blades) are less sensitive. The present CFD pump model has been proved to be an efficient and robust analytical tool for pump design due to its very compact numerical structure (requiring small memory), fast turnaround computing time, and versatility for different geometries.

Cheng, Gary C.↗
DOE OSTI2018

Real-time data analysis for medical diagnosis using FPGA-accelerated neural networks

Real-time analysis of patient data during medical procedures can provide vital diagnostic feedback that significantly improves chances of success. With sensors becoming increasingly fast, frameworks such as Deep Neural Networks are required to perform calculations within the strict timing constraints for real-time operation. However, traditional computing platforms responsible for running these algorithms incur a large overhead due to communication protocols, memory accesses, and static (often generic) architectures. In this work, we implement a low-latency Multi-Layer Perceptron (MLP) processor using Field Programmable Gate Arrays (FPGAs). Unlike CPUs and Graphics Processing Units (GPUs), our FPGA-based design can directly interface sensors, storage devices, display devices and even actuators, thus reducing the delays of data movement between ports and compute pipelines. Moreover, the compute pipelines themselves are tailored specifically to the application, improving resource utilization and reducing idle cycles. We demonstrate the effectiveness of our approach using mass-spectrometry data sets for real-time cancer detection. We demonstrate that correct parameter sizing, based on the application, can reduce latency by 20% on average. Furthermore, we show that in an application with tightly coupled data-path and latency constraints, having a large amount of computing resources can actually reduce performance. Using mass-spectrometry benchmarks, we show that our proposed FPGA design outperforms both CPU and GPU implementations, with an average speedup of 144x and 21x, respectively. In our work, we demonstrate the importance of application-specific optimizations in order to minimize latency and maximize resource utilization for MLP inference. By directly interfacing and processing sensor data with ultra-low latency, FPGAs can perform real-time analysis during procedures and provide diagnostic feedback that can be critical to achieving higher percentages of successful patient outcomes.

60 APPLIED LIFE SCIENCES↗
NASA NTRS2000

Use Computer-Aided Tools to Parallelize Large CFD Applications

Porting applications to high performance parallel computers is always a challenging task. It is time consuming and costly. With rapid progressing in hardware architectures and increasing complexity of real applications in recent years, the problem becomes even more sever. Today, scalability and high performance are mostly involving handwritten parallel programs using message-passing libraries (e.g. MPI). However, this process is very difficult and often error-prone. The recent reemergence of shared memory parallel (SMP) architectures, such as the cache coherent Non-Uniform Memory Access (ccNUMA) architecture used in the SGI Origin 2000, show good prospects for scaling beyond hundreds of processors. Programming on an SMP is simplified by working in a globally accessible address space. The user can supply compiler directives, such as OpenMP, to parallelize the code. As an industry standard for portable implementation of parallel programs for SMPs, OpenMP is a set of compiler directives and callable runtime library routines that extend Fortran, C and C++ to express shared memory parallelism. It promises an incremental path for parallel conversion of existing software, as well as scalability and performance for a complete rewrite or an entirely new development. Perhaps the main disadvantage of programming with directives is that inserted directives may not necessarily enhance performance. In the worst cases, it can create erroneous results. While vendors have provided tools to perform error-checking and profiling, automation in directive insertion is very limited and often failed on large programs, primarily due to the lack of a thorough enough data dependence analysis. To overcome the deficiency, we have developed a toolkit, CAPO, to automatically insert OpenMP directives in Fortran programs and apply certain degrees of optimization. CAPO is aimed at taking advantage of detailed inter-procedural dependence analysis provided by CAPTools, developed by the University of Greenwich, to reduce potential errors made by users. Earlier tests on NAS Benchmarks and ARC3D have demonstrated good success of this tool. In this study, we have applied CAPO to parallelize three large applications in the area of computational fluid dynamics (CFD): OVERFLOW, TLNS3D and INS3D. These codes are widely used for solving Navier-Stokes equations with complicated boundary conditions and turbulence model in multiple zones. Each one comprises of from 50K to 1,00k lines of FORTRAN77. As an example, CAPO took 77 hours to complete the data dependence analysis of OVERFLOW on a workstation (SGI, 175MHz, R10K processor). A fair amount of effort was spent on correcting false dependencies due to lack of necessary knowledge during the analysis. Even so, CAPO provides an easy way for user to interact with the parallelization process. The OpenMP version was generated within a day after the analysis was completed. Due to sequential algorithms involved, code sections in TLNS3D and INS3D need to be restructured by hand to produce more efficient parallel codes. An included figure shows preliminary test results of the generated OVERFLOW with several test cases in single zone. The MPI data points for the small test case were taken from a handcoded MPI version. As we can see, CAPO's version has achieved 18 fold speed up on 32 nodes of the SGI O2K. For the small test case, it outperformed the MPI version. These results are very encouraging, but further work is needed. For example, although CAPO attempts to place directives on the outer- most parallel loops in an interprocedural framework, it does not insert directives based on the best manual strategy. In particular, it lacks the support of parallelization at the multi-zone level. Future work will emphasize on the development of methodology to work in a multi-zone level and with a hybrid approach. Development of tools to perform more complicated code transformation is also needed.

Jin, H.↗
NASA NTRS2011

Interplanetary Overlay Network Bundle Protocol Implementation

The Interplanetary Overlay Network (ION) system's BP package, an implementation of the Delay-Tolerant Networking (DTN) Bundle Protocol (BP) and supporting services, has been specifically designed to be suitable for use on deep-space robotic vehicles. Although the ION BP implementation is unique in its use of zero-copy objects for high performance, and in its use of resource-sensitive rate control, it is fully interoperable with other implementations of the BP specification (Internet RFC 5050). The ION BP implementation is built using the same software infrastructure that underlies the implementation of the CCSDS (Consultative Committee for Space Data Systems) File Delivery Protocol (CFDP) built into the flight software of Deep Impact. It is designed to minimize resource consumption, while maximizing operational robustness. For example, no dynamic allocation of system memory is required. Like all the other ION packages, ION's BP implementation is designed to port readily between Linux and Solaris (for easy development and for ground system operations) and VxWorks (for flight systems operations). The exact same source code is exercised in both environments. Initially included in the ION BP implementations are the following: libraries of functions used in constructing bundle forwarders and convergence-layer (CL) input and output adapters; a simple prototype bundle forwarder and associated CL adapters designed to run over an IPbased local area network; administrative tools for managing a simple DTN infrastructure built from these components; a background daemon process that silently destroys bundles whose time-to-live intervals have expired; a library of functions exposed to applications, enabling them to issue and receive data encapsulated in DTN bundles; and some simple applications that can be used for system checkout and benchmarking.

Burleigh, Scott C.↗
DOE OSTI2024

Developing And Scaling an OpenFOAM Model to Study Turbulent Flow in a HFIR Coolant Channel

Improving the understanding of how computational fluid dynamics (CFD) direct numerical simulations (DNS) of flows in the High Flux Isotope Reactor (HFIR) perform when run in parallel using the high performance computing (HPC) platform Summit at the Oak Ridge Leadership Computing Facility (OLCF) is of particular importance to boost the computational tools used to support HFIR conversion to low enriched fuel (LEU). Evaluation of scaling performance was driven by the increasing importance of graphics processing unit (GPU) usage in HPC, which is becoming the standard for modern supercomputers such as Summit. The desired results are to obtain a strong positive correlation between the computational resources dedicated to a problem and the relative speed-up of the simulation in comparison to a benchmark. This capability will allow substantially improvement in HFIR flow analytical capabilities, specifically when predicting turbulence properties at high Reynolds numbers. The study leverages previous simulation results performed with code PHASTA (finite element) on HPC platforms Cori (NERSC) and Theta (ALCF) [1] with computing options provided in the computing platform OpenFOAM (finite volume) at OLCF. Transitioning from PHASTA to OpenFOAM will (1) eliminate dependence on third-party software for mesh generation and manipulation, (2) reduce resource needs by employing modern architectures, and (3) build expertise for future modeling of HFIR-specific problems like heat transfer in involute geometry, entrance effects, flow structure in channel corners, and so on—all important issues when defining the available thermal margins in the transition to LEU. CPUs and GPUs differ significantly in their architecture and utilization, as discussed in the literature [2]. The most important differences are in the approach to computations and their memory. A single GPU contains a large quantity of cores, enabling it to perform with a much higher throughput than a CPU, but execution requires a different approach. GPU codes execute instructions using the Single-Instruction Multiple-Thread (SIMT) approach in which a single instruction is used for groups of threads called warps. A warp typically consists of 32 threads which must execute the same set of instructions, although on separate threads. Alternately, a CPU has far fewer cores that are much more flexible in their operation, excelling at quickly performing more complex serial computations. This is why GPUs have greater throughput when properly utilized. The second important difference is seen when comparing their memory spaces. Limited memory allocations and CPU–GPU communications cause a significant bottleneck in GPU-accelerated programs. Further study was required to properly take advantage of GPU resources. A comprehensive analysis of code performance and the model-specific features of turbulence constitutes the core of this work. In this study, a DNS simulation of HFIR channel turbulence was performed with the finite volume CFD code OpenFOAM v2112 and CUDA v11.0 on Red Hat Enterprise Linux v8.2. The OpenFOAM installation had AMGx integrated to enable GPU acceleration and utilizes the PETSc4FOAM library. The computational resources and the problem size were scaled on CPU and CPU + GPU architectures to gain a better understanding of the performance of a DNS problem on modern computing hardware. The study aimed to analyze the scaling of the code exclusively on CPUs and then to examine the scaling of the codes with GPU acceleration enabled. Scaling studies included CPU and GPU acceleration on a mesh of varying resolution to analyze the impact of problem size relative to computational resources. In the course of preparing the GPU configuration on Summit, mainly using the AMGX solvers, difficulties were encountered stemming from constant changes resulting from extensive ongoing development activities and the changing environment. This resulted in the inability to complete the GPU portion of the work. The code was compiled and tested, but production runs to assess acceleration were not performed because the used discretional compute time allocation expired as year-end approached. The Summit HPC platform is scheduled for decommissioning in 2024, making it unattractive for future use with Nvidia-based GPUs. Therefore, the work will be moved onto NERSC machines in FY24. An application was prepared and submitted, and sufficient node-hours were awarded to continue the research in the next calendar year. This report summarizes work performed thus far, which mostly focused on CPU OpenFOAM computing.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2018

Reduced scaling CASPT2 using supporting subspaces and tensor hyper-contraction

Here, we present a reduced scaling formulation of the state specific complete active space second-order perturbation method (CASPT2) requiring O( N 4 ) operations and O( N 2 ) memory for a fixed active space, where N is proportional to system size. Motivated by the properties of the Kronecker sum, we introduce the supporting subspace technique (SST), which decomposes the CASPT2 linear equations into two parts: a single-reference MP2 energy term using dressed orbitals, plus a reduced linear system with dimension scaling as O( N 2 ). Together with Laplace quadrature, the SST allows us to reformulate CASPT2 using a MP2 energy computation and Fock builds. By further applying the tensor hyper-contraction (THC) approximation, the MP2-like term can be computed with O( N 4 ) operations, and the remainder can be solved with O( N 3 ) operations using the preconditioned conjugate gradient method. This is the first application of THC in the context of multi-reference methods. We also developed an efficient implementation of the method by utilizing graphical processing units and exploiting spatial sparsity in tensor operations. We benchmark the accuracy of the new method against conventional CASPT2 for reactions in the gas phase. We apply the new method to Menshutkin S N 2 reactions in carbon nanotubes, demonstrating the feasibility of CASPT2 calculations with O(100) atoms.

74 ATOMIC AND MOLECULAR PHYSICS↗
DOE OSTI2019

RoCE: Promising Technology for Ethernet as a High Performance Networking Fabric

Remote Direct Memory Access (RDMA) over Converged Ethernet (RoCE) has the potential to provide performance that rivals traditional high performance fabrics. If this potential proves out, significant impacts on system procurement decisions could follow. This work provides a series of small scale performance results which are used to compare and contrast the performance of RoCE-enabled Ethernet with TCP-based Ethernet and an HPC network. Additionally, a discussion of the maturity of RoCE firmware/software stacks and documentation is provided along with useful approaches for probing performance. A detailed description of two experimental setups known to have good RoCE performance is given, including step-by-step configuration and the exact hardware and software revisions employed. At small scales, RoCE is found to have significant performance advantages over "out-of-the-box" TCP protocols and is competitive with state-of-the-art high performance networks. Further examination of RoCE using a wider array of benchmarks and at greater scale is warranted.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

A Privacy Preserving Model-Free Optimization and Control Framework for Demand Response from Residential Thermal Loads

We consider the problem of optimizing the cost of procuring electricity for a large collection of homes managed by a load serving entity, by pre-cooling or pre-heating the thermal inertial loads in the homes to avoid procuring power during periods of peak electricity pricing. We would like to accomplish this objective in a completely privacy-preserving and model-free manner, that is, without direct access to the state variables (temperatures or power consumption) or the dynamical models (thermal characteristics) of individual homes, while guaranteeing personal comfort constraints of the consumers. We propose a two-stage optimization and control framework to address this problem. In the first stage, we use a long short-term memory (LSTM) network to predict hourly electricity prices, based on historical pricing data and weather forecasts. Given the hourly price forecast and thermal models of the homes, the problem of designing an optimal power consumption trajectory that minimizes the total electricity procurement cost for the collection of thermal loads can be formulated as a large-scale integer program (with millions of variables) due to the on-off cyclical dynamics of such loads. We provide a simple heuristic relaxation to make this large-scale optimization problem model-free and computationally tractable. In the second stage, we translate the results of this optimization problem into distributed open-loop control laws that can be implemented at individual homes without measuring or estimating their state variables, while simultaneously ensuring consumer comfort constraints. We demonstrate the performance of this approach on a large-scale test case comprising of 500 homes in the Houston area and benchmark its performance against a direct model-based optimization and control solution.

Sivaranjani, S.↗
DOE OSTI2023

XploreNAS : Explore Adversarially Robust and Hardware-efficient Neural Architectures for Non-ideal Xbars

Compute In-Memory platforms such as memristive crossbars are gaining focus as they facilitate acceleration of Deep Neural Networks (DNNs) with high area and compute efficiencies. However, the intrinsic non-idealities associated with the analog nature of computing in crossbars limits the performance of the deployed DNNs. Furthermore, DNNs are shown to be vulnerable to adversarial attacks leading to severe security threats in their large-scale deployment. Thus, finding adversarially robust DNN architectures for non-ideal crossbars is critical to the safe and secure deployment of DNNs on the edge. This work proposes a two-phase algorithm-hardware co-optimization approach called XploreNAS that searches for hardware efficient and adversarially robust neural architectures for non-ideal crossbar platforms. We use the one-shot Neural Architecture Search approach to train a large Supernet with crossbar-awareness and sample adversarially robust Subnets therefrom, maintaining competitive hardware efficiency. Our experiments on crossbars with benchmark datasets (SVHN, CIFAR10, CIFAR100) show up to ~8–16% improvement in the adversarial robustness of the searched Subnets against a baseline ResNet-18 model subjected to crossbar-aware adversarial training. We benchmark our robust Subnets for Energy-Delay-Area-Products (EDAPs) using the Neurosim tool and find that with additional hardware efficiency–driven optimizations, the Subnets attain ~1.5–1.6× lower EDAPs than ResNet-18 baseline.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1994

Automated Instrumentation, Monitoring and Visualization of PVM Programs Using AIMS

We present views and analysis of the execution of several PVM (Parallel Virtual Machine) codes for Computational Fluid Dynamics on a networks of Sparcstations, including: (1) NAS Parallel Benchmarks CG and MG; (2) a multi-partitioning algorithm for NAS Parallel Benchmark SP; and (3) an overset grid flowsolver. These views and analysis were obtained using our Automated Instrumentation and Monitoring System (AIMS) version 3.0, a toolkit for debugging the performance of PVM programs. We will describe the architecture, operation and application of AIMS. The AIMS toolkit contains: (1) Xinstrument, which can automatically instrument various computational and communication constructs in message-passing parallel programs; (2) Monitor, a library of runtime trace-collection routines; (3) VK (Visual Kernel), an execution-animation tool with source-code clickback; and (4) Tally, a tool for statistical analysis of execution profiles. Currently, Xinstrument can handle C and Fortran 77 programs using PVM 3.2.x; Monitor has been implemented and tested on Sun 4 systems running SunOS 4.1.2; and VK uses XIIR5 and Motif 1.2. Data and views obtained using AIMS clearly illustrate several characteristic features of executing parallel programs on networked workstations: (1) the impact of long message latencies; (2) the impact of multiprogramming overheads and associated load imbalance; (3) cache and virtual-memory effects; and (4) significant skews between workstation clocks. Interestingly, AIMS can compensate for constant skew (zero drift) by calibrating the skew between a parent and its spawned children. In addition, AIMS' skew-compensation algorithm can adjust timestamps in a way that eliminates physically impossible communications (e.g., messages going backwards in time). Our current efforts are directed toward creating new views to explain the observed performance of PVM programs. Some of the features planned for the near future include: (1) ConfigView, showing the physical topology of the virtual machine, inferred using specially formatted IP (Internet Protocol) packets: and (2) LoadView, synchronous animation of PVM-program execution and resource-utilization patterns.

Mehra, Pankaj↗
NASA NTRS1994

Automated Instrumentation, Monitoring and Visualization of PVM Programs Using AIMS

We present views and analysis of the execution of several PVM codes for Computational Fluid Dynamics on a network of Sparcstations, including (a) NAS Parallel benchmarks CG and MG (White, Alund and Sunderam 1993); (b) a multi-partitioning algorithm for NAS Parallel Benchmark SP (Wijngaart 1993); and (c) an overset grid flowsolver (Smith 1993). These views and analysis were obtained using our Automated Instrumentation and Monitoring System (AIMS) version 3.0, a toolkit for debugging the performance of PVM programs. We will describe the architecture, operation and application of AIMS. The AIMS toolkit contains (a) Xinstrument, which can automatically instrument various computational and communication constructs in message-passing parallel programs; (b) Monitor, a library of run-time trace-collection routines; (c) VK (Visual Kernel), an execution-animation tool with source-code clickback; and (d) Tally, a tool for statistical analysis of execution profiles. Currently, Xinstrument can handle C and Fortran77 programs using PVM 3.2.x; Monitor has been implemented and tested on Sun 4 systems running SunOS 4.1.2; and VK uses X11R5 and Motif 1.2. Data and views obtained using AIMS clearly illustrate several characteristic features of executing parallel programs on networked workstations: (a) the impact of long message latencies; (b) the impact of multiprogramming overheads and associated load imbalance; (c) cache and virtual-memory effects; and (4significant skews between workstation clocks. Interestingly, AIMS can compensate for constant skew (zero drift) by calibrating the skew between a parent and its spawned children. In addition, AIMS' skew-compensation algorithm can adjust timestamps in a way that eliminates physically impossible communications (e.g., messages going backwards in time). Our current efforts are directed toward creating new views to explain the observed performance of PVM programs. Some of the features planned for the near future include: (a) ConfigView, showing the physical topology of the virtual machine, inferred using specially formatted IP (Internet Protocol) packets; and (b) LoadView, synchronous animation of PVM-program execution and resource-utilization patterns.

Mehra, Pankaj↗
NASA NTRS2001

Development of a Hybrid RANS/LES Method for Turbulent Mixing Layers

Significant research has been underway for several years in NASA Glenn Research Center's nozzle branch to develop advanced computational methods for simulating turbulent flows in exhaust nozzles. The primary efforts of this research have concentrated on improving our ability to calculate the turbulent mixing layers that dominate flows both in the exhaust systems of modern-day aircraft and in those of hypersonic vehicles under development. As part of these efforts, a hybrid numerical method was recently developed to simulate such turbulent mixing layers. The method developed here is intended for configurations in which a dominant structural feature provides an unsteady mechanism to drive the turbulent development in the mixing layer. Interest in Large Eddy Simulation (LES) methods have increased in recent years, but applying an LES method to calculate the wide range of turbulent scales from small eddies in the wall-bounded regions to large eddies in the mixing region is not yet possible with current computers. As a result, the hybrid method developed here uses a Reynolds-averaged Navier-Stokes (RANS) procedure to calculate wall-bounded regions entering a mixing section and uses a LES procedure to calculate the mixing-dominated regions. A numerical technique was developed to enable the use of the hybrid RANS-LES method on stretched, non-Cartesian grids. With this technique, closure for the RANS equations is obtained by using the Cebeci-Smith algebraic turbulence model in conjunction with the wall-function approach of Ota and Goldberg. The LES equations are closed using the Smagorinsky subgrid scale model. Although the function of the Cebeci-Smith model to replace all of the turbulent stresses is quite different from that of the Smagorinsky subgrid model, which only replaces the small subgrid turbulent stresses, both are eddy viscosity models and both are derived at least in part from mixing-length theory. The similar formulation of these two models enables the RANS and LES equations to be solved with a single solution scheme and computational grid. The hybrid RANS-LES method has been applied to a benchmark compressible mixing layer experiment in which two isolated supersonic streams, separated by a splitter plate, provide the flows to a constant-area mixing section. Although the configuration is largely two dimensional in nature, three-dimensional calculations were found to be necessary to enable disturbances to develop in three spatial directions and to transition to turbulence. The flow in the initial part of the mixing section consists of a periodic vortex shedding downstream of the splitter plate trailing edge. This organized vortex shedding then rapidly transitions to a turbulent structure, which is very similar to the flow development observed in the experiments. Although the qualitative nature of the large-scale turbulent development in the entire mixing section is captured well by the LES part of the current hybrid method, further efforts are planned to directly calculate a greater portion of the turbulence spectrum and to limit the subgrid scale modeling to only the very small scales. This will be accomplished by the use of higher accuracy solution schemes and more powerful computers, measured both in speed and memory capabilities.

Georgiadis, Nicholas J.↗
DOE OSTI2025

Direct numerical simulations for hybrid rocket boundary layers: Performance modeling and scaling

This paper presents a comprehensive performance and scaling analysis of direct numerical simulations for reacting boundary layers, focusing on slab burner configurations. Using a PETSc-based finite volume CFD framework, the study evaluates the scalability and computational cost of flow, chemistry, and radiation evaluations across 2D and 3D simulations. Polymethyl methacrylate (PMMA) is the fuel with pure O 2 as the oxidizer, modeled using a detailed chemical kinetics mechanism with 113 species and 660 reactions. A ray-tracing-based radiation solver, designed for distributed memory applications, is implemented to model radiation heat transfer. Parallel scalability is analyzed for the coupled flow, chemistry, and radiation heat transfer processes. Weak and strong scaling studies are conducted on up to 15,000 computational ranks, revealing robust performance when flow cells exceed 200 per rank. Chemistry evaluations dominate the computational cost in large 3D simulations, accounting for approximately 40% of the total runtime, while flow processes contribute around 35%, and radiation solver contributions remain below 10% due to reduced evaluation frequencies. GPU accelerated chemistry evaluation, implemented with Zero-RK, demonstrates significant promise, achieving up to a 4x speedup for workloads exceeding 30,000 cells per GPU. However, diminishing returns are observed for smaller workloads due to CPU-GPU communication overhead. This study identifies key challenges, including memory bottlenecks and the effects of domain partitioning on flow scalability, while highlighting the potential of GPU-accelerated chemistry to reduce computational costs. In conclusion, these findings provide realizable run configurations for 2D, 3D, and GPU-accelerated cases, offering insights for optimizing reactive flow solvers.

CFD Scalability↗
DOE OSTI2025

Experiences with SYCL on AMD GPUs with Kokkos

With the recent diversification of the hardware landscape in the high-performance computing (HPC) community, performance-portability solutions are becoming more and more important. One of the most popular choices is Kokkos, which recently became a Linux Foundation project. Most of its development is supported by the US Department of Energy and the French Alternative Energies and Atomic Energy Commission. Kokkos is implemented as a C++ library with multiple backends to support CPUs as well as various GPU architectures. These backends include OpenMP, CUDA, HIP, and also SCYL. This approach enables users to leverage the preferred vendor toolchain for the respective platform (e.g. CUDA, ROCm, OneAPI). The SYCL backend is used to target Intel GPUs, in particular to support the Aurora exascale supercomputer. However, SYCL itself also offers a large degree of portability, and in fact Kokkos’ CI for SYCL has been running on NVIDIA hardware due to a lack of access to Intel GPUs. In this report, we describe our experience with using Kokkos SYCL backend on AMD GPUs targeting the Frontier supercomputer at Oak Ridge National Laboratory. The two major SYCL implementations are DPC++ and AdaptiveCpp. While the Kokkos SYCL backend has been implemented using the former, the latter was the first implementation to target AMD GPUs. We will discuss the experience with both of these SYCL implementations in terms of functionality and performance. Using Kokkos to evaluate SYCL toolchains has a number of benefits. Kokkos’ use of SYCL is fairly complex, exercising features such as graphs, relocatable device functions, atomics – including for non-arithmetic types, as well as pinned and page migratable memory allocations. Kokkos also needs to implement capabilities such as Kokkos’ hierarchical parallelism that are not a straight-forward mapping to SYCL capabilities. Furthermore, a large number of libraries and applications that represent diverse use cases are implemented in Kokkos, providing readily available test cases for a toolchain evaluation. Preliminary results show that support for AMD GPUs in DPC++ is much less mature than for NVIDIA GPUs or Intel GPUs. While the situation has improved significantly over the last year, we still encounter many runtime failures, dispatching problems, and code generation issues. With AdaptiveCpp the challenges arise even earlier in the evaluation process. Since Kokkos’ SYCL implementation is largely focused on supporting Intel GPUs, we opted to leverage SYCL extensions which are available in DPC++ but not in AdaptiveCpp. Furthermore, AdaptiveCpp appears to be less conformant with the SYCL2020 standard which Kokkos relies on. In some cases, we are able to work around the lack of feature support, in other cases we have to disable certain Kokkos capabilities to evaluate the toolchain. Our evaluation will leverage Kokkos’ unit tests to establish basic functionality and feature completeness. We then use simple benchmarks for components of a CG implementation as a measure of usability and performance of the SYCL toolchains.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Performance improvements of the windowed multipole formalism using a rational fraction approximation of the Faddeeva function

The windowed multipole (WMP) formalism was introduced as a way to calculate Doppler broadened cross sections on the fly during Monte Carlo simulations. While more arithmetic is needed compared to point-wise cross section look-ups, performance remained competitive from the large memory reductions and sequential data access. The single most expensive function call in a depleted fuel assembly problem using WMP comes from the evaluation of the Faddeeva function, which previously relied on a highly accurate, highly-branching algorithm. This paper explores the use of rational fraction approximations tailored to the domain interest of reactor physics applications and the development of lower accuracy approximations sufficient for our application. The rational approximations were implemented and tested in OpenMC on an infinite medium problem to stress the cross section calculation routine and a PWR assembly problem. In both cases, the rational approximation nearly eliminated the ∼ 20% penalty previously observed when comparing to point-wise libraries. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

A graphics processing unit accelerated sparse direct solver and preconditioner with block low rank compression

We present the GPU implementation efforts and challenges of the sparse solver package STRUMPACK. The code is made publicly available on github with a permissive BSD license. STRUMPACK implements an approximate multifrontal solver, a sparse LU factorization which makes use of compression methods to accelerate time to solution and reduce memory usage. Multiple compression schemes based on rank-structured and hierarchical matrix approximations are supported, including hierarchically semi-separable, hierarchically off-diagonal butterfly, and block low rank. Here, in this paper, we present the GPU implementation of the block low rank (BLR) compression method within a multifrontal solver. Our GPU implementation relies on highly optimized vendor libraries such as cuBLAS and cuSOLVER for NVIDIA GPUs, rocBLAS and rocSOLVER for AMD GPUs and the Intel oneAPI Math Kernel Library (oneMKL) for Intel GPUs. Additionally, we rely on external open source libraries such as SLATE (Software for Linear Algebra Targeting Exascale), MAGMA (Matrix Algebra on GPU and Multi-core Architectures), and KBLAS (KAUST BLAS). SLATE is used as a GPU-capable ScaLAPACK replacement. From MAGMA we use variable sized batched dense linear algebra operations such as GEMM, TRSM and LU with partial pivoting. KBLAS provides efficient (batched) low rank matrix compression for NVIDIA GPUs using an adaptive randomized sampling scheme. The resulting sparse solver and preconditioner runs on NVIDIA, AMD and Intel GPUs. Interfaces are available from PETSc, Trilinos and MFEM, or the solver can be used directly in user code. We report results for a range of benchmark applications, using the Perlmutter system from NERSC, Frontier from ORNL, and Aurora from ALCF. For a high frequency wave equation on a regular mesh, using 32 Perlmutter compute nodes, the factorization phase of the exact GPU solver is about 6.5× faster compared to the CPU-only solver. The BLR-enabled GPU solver is about 13.8× faster than the CPU exact solver. For a collection of SuiteSparse matrices, the STRUMPACK exact factorization on a single GPU is on average 1.9× faster than NVIDIA’s cuDSS solver.

97 MATHEMATICS AND COMPUTING↗