SEARCH · Search NASA

Results for “OVERFLOW”

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.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 19 records

All sources NASA NTRS DOE OSTI

DOE OSTI2025

Role of the junction voltage on the overflow current in light-emitting diodes

Quantum-well (QW)-based light emitters, such as light-emitting diodes (LEDs) and lasers, of various semiconductor materials experience a reduction in their efficiency when operating at higher temperatures, a phenomenon referred to as “thermal droop.” Among the various claims on the origins of thermal droop, an increased overflow current with increasing temperatures is a common contender. Since overflow of carriers can only occur when the junction voltage 𝑉 Junction approaches the built-in voltage 𝑉 BI of any diodes, we develop a simple method relating the difference between 𝑉 Junction and 𝑉 BI to approximate the upper limit of overflow occurring in QW-based light-emitting diodes. The measured difference between 𝑉 Junction and 𝑉 BI of state-of-the-art commercial blue and green In⁢Ga⁢N-based LEDs at temperatures up to ∼450 K suggests negligible overflow. To further experimentally verify the absence of overflow, we perform temperature-dependent electron emission spectroscopy on the same commercial blue and green LEDs and find no evidence of thermally enhanced overflow carriers up to ∼450 K. In agreement with our claims that 𝑉 Junction must approach 𝑉 BI for overflow to occur, two-dimensional temperature-dependent electrical simulations of violet, blue, and green LEDs including alloy disorder and V-defects demonstrate that overflow can be significant in violet LEDs, where the small band offset between the In⁢Ga⁢N QW and Ga⁢N cladding layers due to the larger QW bandgap requires larger 𝑉 Junction to reach standard operating current densities, thereby approaching 𝑉 BI . By contrast, simulations indicate that overflow is negligible in blue and green LEDs, whose smaller QW bandgaps result in smaller quasi-Fermi levels difference to reach significant carrier injection, resulting in a 𝑉 Junction much smaller than 𝑉 BI up to large operating current densities. Considering that overflow is negligible in blue and longer-wavelength LEDs, and our observations of the large thermal droop occurring at low current densities, where Shockley-Read-Hall (SRH) recombination dominates, we conclude that thermally enhanced SRH processes are the most significant contributor to thermal droop. Finally, we also simulate the carrier densities in the different QWs of a multiple-QW LED and observe a reduction in the total carrier density at a given operating current density, which results in a decrease in the total Auger-Meitner current of the LED from just the thermally enhanced carrier redistribution among QWs without taking any possible additional temperature dependence of their recombination coefficients. Taking all this into account, minimizing thermal droop effects in LEDs can be achieved by a reduction in defect density, using wider band gap p-n junction-defining cladding layers, and operating at higher currents.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

NASA NTRS1999

Parallelizing OVERFLOW: Experiences, Lessons, Results

The computer code OVERFLOW is widely used in the aerodynamic community for the numerical solution of the Navier-Stokes equations. Current trends in computer systems and architectures are toward multiple processors and parallelism, including distributed memory. This report describes work that has been carried out by the author and others at Ames Research Center with the goal of parallelizing OVERFLOW using a variety of parallel architectures and parallelization strategies. This paper begins with a brief description of the OVERFLOW code. This description includes the basic numerical algorithm and some software engineering considerations. Next comes a description of a parallel version of OVERFLOW, OVERFLOW/PVM, using PVM (Parallel Virtual Machine). This parallel version of OVERFLOW uses the manager/worker style and is part of the standard OVERFLOW distribution. Then comes a description of a parallel version of OVERFLOW, OVERFLOW/MPI, using MPI (Message Passing Interface). This parallel version of OVERFLOW uses the SPMD (Single Program Multiple Data) style. Finally comes a discussion of alternatives to explicit message-passing in the context of parallelizing OVERFLOW.

Jespersen, Dennis C.↗

NASA NTRS2000

Performance of the OVERFLOW-MLP and LAURA-MLP CFD Codes on the NASA Ames 512 CPU Origin System

The shared memory Multi-Level Parallelism (MLP) technique, developed last year at NASA Ames has been very successful in dramatically improving the performance of important NASA CFD codes. This new and very simple parallel programming technique was first inserted into the OVERFLOW production CFD code in FY 1998. The OVERFLOW-MLP code's parallel performance scaled linearly to 256 CPUs on the NASA Ames 256 CPU Origin 2000 system (steger). Overall performance exceeded 20.1 GFLOP/s, or about 4.5x the performance of a dedicated 16 CPU C90 system. All of this was achieved without any major modification to the original vector based code. The OVERFLOW-MLP code is now in production on the inhouse Origin systems as well as being used offsite at commercial aerospace companies. Partially as a result of this work, NASA Ames has purchased a new 512 CPU Origin 2000 system to further test the limits of parallel performance for NASA codes of interest. This paper presents the performance obtained from the latest optimization efforts on this machine for the LAURA-MLP and OVERFLOW-MLP codes. The Langley Aerothermodynamics Upwind Relaxation Algorithm (LAURA) code is a key simulation tool in the development of the next generation shuttle, interplanetary reentry vehicles, and nearly all "X" plane development. This code sustains about 4-5 GFLOP/s on a dedicated 16 CPU C90. At this rate, expected workloads would require over 100 C90 CPU years of computing over the next few calendar years. It is not feasible to expect that this would be affordable or available to the user community. Dramatic performance gains on cheaper systems are needed. This code is expected to be perhaps the largest consumer of NASA Ames compute cycles per run in the coming year.The OVERFLOW CFD code is extensively used in the government and commercial aerospace communities to evaluate new aircraft designs. It is one of the largest consumers of NASA supercomputing cycles and large simulations of highly resolved full aircraft are routinely undertaken. Typical large problems might require 100s of Cray C90 CPU hours to complete. The dramatic performance gains with the 256 CPU steger system are exciting. Obtaining results in hours instead of months is revolutionizing the way in which aircraft manufacturers are looking at future aircraft simulation work. Figure 2 below is a current state of the art plot of OVERFLOW-MLP performance on the 512 CPU Lomax system. As can be seen, the chart indicates that OVERFLOW-MLP continues to scale linearly with CPU count up to 512 CPUs on a large 35 million point full aircraft RANS simulation. At this point performance is such that a fully converged simulation of 2500 time steps is completed in less than 2 hours of elapsed time. Further work over the next few weeks will improve the performance of this code even further.The LAURA code has been converted to the MLP format as well. This code is currently being optimized for the 512 CPU system. Performance statistics indicate that the goal of 100 GFLOP/s will be achieved by year's end. This amounts to 20x the 16 CPU C90 result and strongly demonstrates the viability of the new parallel systems rapidly solving very large simulations in a production environment.

Taft, James R.↗

NASA NTRS1999

Isolated Nozzle Analysis Using OVERFLOW

LaRC conducted a code validation study for the OVERFLOW code to ascertain its accuracy for boattail drag prediction. The OVERFLOW results compared favorably with the LaRC 16-ft. Transonic Wind Tunnel (TWT) data, and prior CFD solutions from PAB3D and CFL3D. The ultimate goal is to investigate the installation drag of the nacelle boattails with powered nozzles at transonic mach numbers. The OVERFLOW solver was chosen because of its ability to accept volume overlapping structured grid for very complex airframe configurations. Structured grid components for representing the transonic nozzle boattail can be added to the BCAG grid for a TCA airframe with 2D bifurcated inlet and flow through nacelle without alteration. The focus of this research was to determine the suitability of the OVERFLOW solver for accomplishing this ultimate goal. This presentation will first introduce the transonic nozzle boattail wind-tunnel model geometry, followed by an examination of aerodynamic features based on the current OVERFLOW solutions and the solutions obtained previously using PAB3D, comparisons of Cp on the flap surface between the OVERFLOW solutions, wind tunnel data, and solutions from other CFD codes, an assessment of boattail drag count prediction, and a work plan for FY99.

Wong, Tin-Chee↗

NASA NTRSDate not supplied

Exploring Airfoil Table Generation using XFOIL and OVERFLOW

The rotorcraft design process is a continuously evolving field of research that incorporates a number of software programs. An accurate airfoil table is critical in the design and testing process for rotorcraft. With multiple flow solvers available and flow conditions of multirotor UAM vehicles potentially covering a wide range of Reynolds and Mach numbers, a documented approach for developing airfoil tables is needed. Using benchmark data from legacy airfoil tables and wind tunnel tests for comparison, simulations for a comprehensive test matrix could guide rotorcraft design engineers in generating their own airfoil tables using the XFOIL and OVERFLOW solvers. The motivation for this study is to investigate flow solver features to develop a best practices document for airfoil table generation. The study uses the OVERFLOW and XFOIL flow solvers, coupled with the airfoil table generator AFTGen, to analyze three airfoils for a specific Reynolds numbers flow regime and provide details on how well each flow solver performs within a specific angle of attack range, Mach number range, Reynolds number range, and in different flow conditions, such as turbulent and transitional flow. OVERFLOW analyses in AFTGen for fully turbulent and transition flow are compared with XFOIL results and experimental test data for the section lift, section drag, and pressure coefficients. XFOIL ultimately yields results that are accurate within the linear angle of attack range and below a Mach number of 0.4 but tends to overpredict lift and underpredict drag unless the flow is in the compressible regime. XFOIL cannot accurately model stall and post-stall conditions due to the nature of the solver. This is evident in nearly every case run with XFOIL, where the linear range is usually predicted acceptably and the lift coefficient is overpredicted as the stall angle of attack is approached (with the exception being the generally poor correlation with most of the SSC-A09 cases). OVERFLOW is limited at low Mach numbers, and appears to perform best at Mach numbers of 0.4 and above. The exploration of airfoil table generation using XFOIL and OVERFLOW yielded moderately successful results for the NACA 0012 airfoil, reasonably good results for the RC(4)-10 airfoil, and less accurate results for the SSC-A09 airfoil.

Airfoil Table Generation↗

NASA NTRSDate not supplied

PALMO: An OVERFLOW Machine Learning Airfoil Performance Database

The OVERFLOW Machine Learning Airfoil Performance (PALMO) database has been created to enable robust modeling of airfoil performance in a variety of applications. The PALMO database uses OVERFLOW simulation data second-order accurate in time and fourth-order accurate in space with Spalart-Allmaras turbulence closure. The foundation of the in-development PALMO database is the airfoil base cube. Each base cube includes simulation data parametrized over a range of Mach numbers, Reynolds numbers, and angles-of-attack. This database includes the NACA 4-series airfoils, with parametrization in airfoil thickness and camber from an NACA 0006 to an NACA 4424. In total, 52,480 NACA 4-series OVERFLOW calculations were run on the NASA High-End Compute Capability (HECC) supercomputer. This provides high-order-accurate simulation data covering a wide range of aerospace design applications, which enables users to develop accurate airfoil performance look-up tables without additional high-performance computing. In addition to engineering design and analysis of aerospace vehicles, PALMO is well suited to be a benchmark dataset for the development and testing of machine learning methods in aerospace engineering. This work presents an example PALMO surrogate model that enables accurate airfoil performance predictions for any arbitrary combination of camber, thickness, Mach number, Reynolds number, and angle of attack within the bounds of the database. Airfoil performance tables predicted for an airfoil not used in training the model are used in three-dimensional OVERFLOW simulations to quantify the downstream accuracy on aggregate rotor performance metrics. For the NACA 3415 airfoil, which had no common thickness or camber with the training data, the surrogate predicted and CFD generated tables were within 2.1% of each other in the forward flight lift to drag metric. This suggests that performance tables generated for airfoils within the bounds of the PALMO database will yield aggregate rotor performance predictions on par with tables generated from directly running OVERFLOW airfoil calculations. The PALMO airfoil performance coefficients are available publicly.

Database↗

NASA NTRS2024

PALMO: An OVERFLOW Machine Learning Airfoil Performance Database

The OVERFLOW Machine Learning Airfoil Performance (PALMO) database has been created to enable robust modeling of airfoil performance in a variety of applications. The database uses OVERFLOW simulation data second-order accurate in time and fourth-order accurate in space with Spalart-Allmaras turbulence closure. The foundation of the in-development PALMO database is the airfoil base cube. Each base cube includes simulation data parametrized over a range of Mach numbers, Reynolds numbers, and angles-of-attack. This first release of the database includes the NACA 4-series airfoils, with parametrization in airfoil thickness and camber from an NACA 0006 to an NACA 4424. In total, 52,480 NACA 4-series calculations were run on the NASA High-End Compute Capability (HECC) supercomputer and the corresponding airfoil performance coefficients are embedded in the Appendix of this document for public distribution. This provides high-order-accurate simulation data covering a wide range of aerospace design applications, which enables users to develop OVERFLOW-quality airfoil performance look-up tables without additional high-performance computing. In addition to engineering design and analysis of aerospace vehicles, PALMO is well suited to be a benchmark dataset for the development and testing of machine learning methods in aerospace engineering. Downstream surrogate models enable OVERFLOW- quality airfoil performance predictions for any arbitrary combination of camber, thickness, Mach number, Reynolds number, and angle-of-attack within the bounds of the database.

Database↗

NASA NTRS2016

OVERFLOW Turbulence Modeling Resource Validation Results

Abstract:We exercise the computational fluid dynamics code OVERFLOW on sixteen turbulence model validation cases from the NASALangley Turbulence Model Resource web site. We give some information about the OVERFLOW options used to run these cases, and compare OVERFLOW results with results from other codes and with experiment. The goal is turbulence model validation for OVERFLOW.

computational fluid dynamics↗

NASA NTRSDate not supplied

Porting OVERFLOW CFD Code to GPUs: To Hackathons and Beyond!

OVERFLOW is an overset, structured computational fluid dynamics (CFD) code written in Fortran which is widely used in the government, industry, and academia. Over the last several years the OVERFLOW developers have been working to port miniapps based on computationally expensive parts of OVERFLOW to run on GPUs, primarily using OpenACC. This effort started at our first hackathon in 2019 and since then the OVERFLOW team has attended two additional hackathons (virtually). These hackathon environments have provided a great place to collaborate with others and learn from experts. These learning experiences enabled porting two miniapps to run effectively on NVIDIA GPUs using OpenACC. The first miniapp focused on motifs found in the solver itself and the final ported version runs three times fast ona single V100 compared to a 40 core, dual-socket Intel Skylake node. The speed up in this solverminiapp required multiple design changes including increasing the amount of parallelism available and the amount of work performed in each kernel. The second miniapp focused on overset MPI communication, also saw significant speedups over the CPU implementation using a CUDA-aware MPI implementation through OpenACC. This presentation will discuss our experience at the hackathons, our process of porting the miniapps to run on the GPUs, and several lessons learned throughout.

OpenACC↗

NASA NTRS1999

Performance and Application of Parallel OVERFLOW Codes on Distributed and Shared Memory Platforms

The presentation discusses recent studies on the performance of the two parallel versions of the aerodynamics CFD code, OVERFLOW_MPI and _MLP. Developed at NASA Ames, the serial version, OVERFLOW, is a multidimensional Navier-Stokes flow solver based on overset (Chimera) grid technology. The code has recently been parallelized in two ways. One is based on the explicit message-passing interface (MPI) across processors and uses the _MPI communication package. This approach is primarily suited for distributed memory systems and workstation clusters. The second, termed the multi-level parallel (MLP) method, is simple and uses shared memory for all communications. The _MLP code is suitable on distributed-shared memory systems. For both methods, the message passing takes place across the processors or processes at the advancement of each time step. This procedure is, in effect, the Chimera boundary conditions update, which is done in an explicit "Jacobi" style. In contrast, the update in the serial code is done in more of the "Gauss-Sidel" fashion. The programming efforts for the _MPI code is more complicated than for the _MLP code; the former requires modification of the outer and some inner shells of the serial code, whereas the latter focuses only on the outer shell of the code. The _MPI version offers a great deal of flexibility in distributing grid zones across a specified number of processors in order to achieve load balancing. The approach is capable of partitioning zones across multiple processors or sending each zone and/or cluster of several zones into a single processor. The message passing across the processors consists of Chimera boundary and/or an overlap of "halo" boundary points for each partitioned zone. The MLP version is a new coarse-grain parallel concept at the zonal and intra-zonal levels. A grouping strategy is used to distribute zones into several groups forming sub-processes which will run in parallel. The total volume of grid points in each group are approximately balanced. A proper number of threads are initially allocated to each group, and in subsequent iterations during the run-time, the number of threads are adjusted to achieve load balancing across the processes. Each process exploits the multitasking directives already established in Overflow.

Djomehri, M. Jahed↗

NASA NTRS1999

Overflow: Facts on Friction

This paper presents results of three minor studies into the behavior of the OVERFLOW with respect to the prediction of skin friction drag on wing bodies at cruise Mach number and wind tunnel Reynolds number. The studies include a preliminary assessment of the behavior of the two new 2-equation turbulence models introduced with the latest version of OVERFLOW (v. 1.8f), an investigation into potential improvements in the matrix dissipation scheme currently implemented in OVERFLOW, and an analysis of the observed sensitivity of the code's skin friction predictions to grid stretching at solid surface boundaries.

Lawrence, Scott↗

NASA NTRS2000

Performance Assessment of OVERFLOW on Distributed Computing Environment

The aerodynamic computer code, OVERFLOW, with a multi-zone overset grid feature, has been parallelized to enhance its performance on distributed and shared memory paradigms. Practical application benchmarks have been set to assess the efficiency of code's parallelism on high-performance architectures. The code's performance has also been experimented with in the context of the distributed computing paradigm on distant computer resources using the Information Power Grid (IPG) toolkit, Globus. Two parallel versions of the code, namely OVERFLOW-MPI and -MLP, have developed around the natural coarse grained parallelism inherent in a multi-zonal domain decomposition paradigm. The algorithm invokes a strategy that forms a number of groups, each consisting of a zone, a cluster of zones and/or a partition of a large zone. Each group can be thought of as a process with one or multithreads assigned to it and that all groups run in parallel. The -MPI version of the code uses explicit message-passing based on the standard MPI library for sending and receiving interzonal boundary data across processors. The -MLP version employs no message-passing paradigm; the boundary data is transferred through the shared memory. The -MPI code is suited for both distributed and shared memory architectures, while the -MLP code can only be used on shared memory platforms. The IPG applications are implemented by the -MPI code using the Globus toolkit. While a computational task is distributed across multiple computer resources, the parallelism can be explored on each resource alone. Performance studies are achieved with some practical aerodynamic problems with complex geometries, consisting of 2.5 up to 33 million grid points and a large number of zonal blocks. The computations were executed primarily on SGI Origin 2000 multiprocessors and on the Cray T3E. OVERFLOW's IPG applications are carried out on NASA homogeneous metacomputing machines located at three sites, Ames, Langley and Glenn. Plans for the future will exploit the distributed parallel computing capability on various homogeneous and heterogeneous resources and large scale benchmarks. Alternative IPG toolkits will be used along with sophisticated zonal grouping strategies to minimize the communication time across the computer resources.

Djomehri, M. Jahed↗

NASA NTRS2007

Drag Prediction for the DLR-F6 Wing/Body and DPW Wing using CFL3D and OVERFLOW Overset Mesh

A series of overset grids was generated in response to the 3rd AIAA CFD Drag Prediction Workshop (DPW-III) which preceded the 25th Applied Aerodynamics Conference in June 2006. DPW-III focused on accurate drag prediction for wing/body and wing-alone configurations. The grid series built for each configuration consists of a coarse, medium, fine, and extra-fine mesh. The medium mesh is first constructed using the current state of best practices for overset grid generation. The medium mesh is then coarsened and enhanced by applying a factor of 1.5 to each (I,J,K) dimension. The resulting set of parametrically equivalent grids increase in size by a factor of roughly 3.5 from one level to the next denser level. CFD simulations were performed on the overset grids using two different RANS flow solvers: CFL3D and OVERFLOW. The results were post-processed using Richardson extrapolation to approximate grid converged values of lift, drag, pitching moment, and angle-of-attack at the design condition. This technique appears to work well if the solution does not contain large regions of separated flow (similar to that seen n the DLR-F6 results) and appropriate grid densities are selected. The extra-fine grid data helped to establish asymptotic grid convergence for both the OVERFLOW FX2B wing/body results and the OVERFLOW DPW-W1/W2 wing-alone results. More CFL3D data is needed to establish grid convergence trends. The medium grid was utilized beyond the grid convergence study by running each configuration at several angles-of-attack so drag polars and lift/pitching moment curves could be evaluated. The alpha sweep results are used to compare data across configurations as well as across flow solvers. With the exception of the wing/body drag polar, the two codes compare well qualitatively showing consistent incremental trends and similar wing pressure comparisons.

Sclanfani, Anthony J.↗

NASA NTRS2010

Drag Prediction for the NASA CRM Wing-Body-Tail Using CFL3D and OVERFLOW on an Overset Mesh

In response to the fourth AIAA CFD Drag Prediction Workshop (DPW-IV), the NASA Common Research Model (CRM) wing-body and wing-body-tail configurations are analyzed using the Reynolds-averaged Navier-Stokes (RANS) flow solvers CFL3D and OVERFLOW. Two families of structured, overset grids are built for DPW-IV. Grid Family 1 (GF1) consists of a coarse (7.2 million), medium (16.9 million), fine (56.5 million), and extra-fine (189.4 million) mesh. Grid Family 2 (GF2) is an extension of the first and includes a superfine (714.2 million) and an ultra-fine (2.4 billion) mesh. The medium grid anchors both families with an established build process for accurate cruise drag prediction studies. This base mesh is coarsened and enhanced to form a set of parametrically equivalent grids that increase in size by a factor of roughly 3.4 from one level to the next denser level. Both CFL3D and OVERFLOW are run on GF1 using a consistent numerical approach. Additional OVERFLOW runs are made to study effects of differencing scheme and turbulence model on GF1 and to obtain results for GF2. All CFD results are post-processed using Richardson extrapolation, and approximate grid-converged values of drag are compared. The medium grid is also used to compute a trimmed drag polar for both codes.

Sclafani, Anthony J.↗

NASA NTRS2023

Implementation and Assessment of Menter’s Galilean-Invariant γ Transition Model in OVERFLOW

With an increased emphasis on greener air transports and sustainable aviation, the modeling of laminar-to-turbulent boundary layer transition is anticipated to have an added significance, particularly in the applications related to laminar flow technology. However, unmanned aerial vehicles, crewed reentry vehicles, and ground-to-flight extrapolation all benefit from transition models. Because no single transition model is ideal for the complete spectrum of applications, it is useful to incorporate a variety of models in general-purpose CFD solvers, such as the NASA OVERFLOW Overset CFD code. While the Langtry-Menter 𝛄 − 𝑹𝒆 𝛉𝒕 model, currently available in OVERFLOW, has been widely used for CFD predictions of flows with laminar, transitional, and turbulent boundary layers, it does not possess the Galilean invariance property, a desirable attribute for rotorcraft applications. To help overcome that limitation, we have recently implemented Menter's baseline version of the SST-based γ transition model, along with a Galilean invariant stationary crossflow extension within OVERFLOW (version 2.3e). An initial assessment of the newly implemented model has been carried out using 2D benchmark cases including flat plates and the NLF-0416 airfoil, addressing several transition scenarios ranging from bypass transition due to freestream turbulence, natural transition via Tollmien-Schlichting instabilities, and transition due to a laminar separation bubble. The crossflow extension has been applied to the infinite swept NLF(2)-0415 wing and the 6:1 prolate spheroid. Wherever possible, the results were obtained on a sequence of meshes to ascertain the grid convergence behavior, which has been evaluated through global metrics such as force coefficients as well as local values of the skin-friction coefficient at selected points near and within the transition region. Overall, the model appears to be correctly implemented and the results show promise for further development using the framework of the γ transition model.

CFD modeling↗

NASA NTRSDate not supplied

GPU Implementation of the OVERFLOW CFD Code

The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.

GPU Programming↗

NASA NTRSDate not supplied

GPU Implementation of the OVERFLOW CFD Code

GPU Programming↗

NASA NTRSDate not supplied

Strategies for the GPU Implementation of the OVERFLOW CFD Code

Wondering how to port a large, computational fluid dynamics (CFD) solver, written in Fortran, to run effectively on GPUs? Join this talk to learn about the strategies NASA’s OVERFLOW CFD code has used to effectively utilize GPUs to greatly improve the time to solution compared to CPUs. OVERFLOW is an overset, structured grid, computational fluid dynamics (CFD) flow solver developed by NASA and widely used by government, US industry, and academia. It is known for its effective use of CPU hardware, but this talk will discuss recent efforts to modify the code to run efficiently on GPUs. We will describe our use of OpenACC, CUDA Fortran, and CUDA C++, discussing why and how we use each to map our problem onto NVIDIA GPUs. We will also describe structural changes we made in the code to expose enough parallelism to effectively use the GPU hardware. Finally, we will present the performance benefits from running OVERFLOW on GPUs compared to the well optimized CPU version of the code.

GPU Programming↗