An MPI-CUDA approach for hypersonic flows with detailed state-to-state air kinetics using a GPU cluster

Bonelli, Francesco; Tuttafesta, M.; Colonna, G.; Cutrone, Luigi; Pascazio, G.

doi:10.1016/j.cpc.2017.05.019

Cited by 51 publications

(22 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…During the last few years, the capability of GPU is growing much faster than that of CPU's because of greatly increasing hardware requirement for modern computer games. The GPU is also rapidly and widely used for various scientific computations in addition to graphic display [30], such as fluid dynamics [31], biophysics [32], molecular dynamics [33], and IoT sensing [34]. GPUs can provide huge performance improvement than a single CPU core for many applications.…”

Section: Cuda Gpumentioning

confidence: 99%

See 1 more Smart Citation

On GPU Implementation of the Island Model Genetic Algorithm for Solving the Unequal Area Facility Layout Problem

et al. 2018

View full text Add to dashboard Cite

Facility layout problem (FLP) is one of the hottest research areas in industrial engineering. A good facility layout can achieve efficient production management, improve production efficiency, and create high economic values. Because FLP is an NP-hard problem, meaning it is impossible to find the optimal solution when problem becomes sufficiently large, various evolutionary algorithms (EAs) have been proposed to find a sub-optimal solution within a reasonable time interval. Recently, a genetic algorithm (GA) was proposed for unequal area FLP (UA-FLP), where the areas of facilities are not identical. More precisely, the GA is an island model based, which is called IMGA. Since EAs are still very time consuming, many efforts have been devoted to how to parallelize various EAs including IMGA. In recent work, Steffen and Dietmar proposed how to parallelize island models of EAs. However, their parallelization approaches are preliminary because they focused mainly on comparing the performances between different parallel architectures. In addition, they used one mathematical function to model the problem. To further investigate on how to parallelize the IMGA by GPU, in this paper we propose multiple parallel algorithms, for each individual step in the IMGA when solving the industrial engineering problem, UA-FLP, and conduct experiments to compare their performances. After integrating better algorithms for all steps into the IMGA, our GPU implementation outperforms the CPU counterpart and the best speedup can be as high as 84.

show abstract

Section: Cuda Gpumentioning

confidence: 99%

“…GPUs can provide huge performance improvement than a single CPU core for many applications. For instance, the work in the CFD field [31] has shown speed-up values of a single GPU against a single core CPU larger than 100. 'General-purpose computing on graphics processing unit' (GPGPU) was thus coined [35].…”

Section: Cuda Gpumentioning

confidence: 99%

On GPU Implementation of the Island Model Genetic Algorithm for Solving the Unequal Area Facility Layout Problem

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Multi-GPU Parallelization Based on MPI+CUDA. The Message Passing Interface (MPI) is widely used on shared and distributed memory machines to implement largescale calculations on multi-GPU parallel cluster [5,29,30]. Figure 6 shows the multi-GPU programming model based on MPI+CUDA.…”

Section: Cuda and Gpu Parallel Algorithm For Cfdmentioning

confidence: 99%

“…What is more, it is attractive for engineers and researchers to obtain accurate numerical results as soon as possible. The graphics processing unit (GPU) has exhibited significant achievements in the accelerated solution of hypersonic flow problems [5,6].…”

Section: Introductionmentioning

confidence: 99%

A Multi‐GPU Parallel Algorithm in Hypersonic Flow Computations

Lai

Tian

et al. 2019

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Computational fluid dynamics (CFD) plays an important role in the optimal design of aircraft and the analysis of complex flow mechanisms in the aerospace domain. The graphics processing unit (GPU) has a strong floating-point operation capability and a high memory bandwidth in data parallelism, which brings great opportunities for CFD. A cell-centred finite volume method is applied to solve three-dimensional compressible Navier–Stokes equations on structured meshes with an upwind AUSM+UP numerical scheme for space discretization, and four-stage Runge–Kutta method is used for time discretization. Compute unified device architecture (CUDA) is used as a parallel computing platform and programming model for GPUs, which reduces the complexity of programming. The main purpose of this paper is to design an extremely efficient multi-GPU parallel algorithm based on MPI+CUDA to study the hypersonic flow characteristics. Solutions of hypersonic flow over an aerospace plane model are provided at different Mach numbers. The agreement between numerical computations and experimental measurements is favourable. Acceleration performance of the parallel platform is studied with single GPU, two GPUs, and four GPUs. For single GPU implementation, the speedup reaches 63 for the coarser mesh and 78 for the finest mesh. GPUs are better suited for compute-intensive tasks than traditional CPUs. For multi-GPU parallelization, the speedup of four GPUs reaches 77 for the coarser mesh and 147 for the finest mesh; this is far greater than the acceleration achieved by single GPU and two GPUs. It is prospective to apply the multi-GPU parallel algorithm to hypersonic flow computations.

show abstract

“…Such phenomena may, in the first instance, be studied using simplified numerical tools [11]. Other phenomena of importance include the possible effects that thermochemical non-equilibrium has on the shock location and therefore on the performance of the inlet compression system, which were outside of the scope of this work [12].…”

Section: Configuration 1-mach 25 Analysismentioning

confidence: 99%

Ramjet Compression System for a Hypersonic Air Transportation Vehicle Combined Cycle Engine

2018

View full text Add to dashboard Cite

This report assesses the performance characteristics of a ramjet compression system in the application of a hypersonic vehicle. The vehicle is required to be self-powered and perform a complete flight profile using a combination of turbojet, ramjet and scramjet propulsion systems. The ramjet has been designed to operate between Mach 2.5 to Mach 5 conditions, allowing for start-up of the scramjet engine. Multiple designs, including varying ramp configurations and turbo-ramjet combinations, were investigated to evaluate their merits and limitations. Challenges arose with attempting to maintain sufficient pressure recoveries and favourable flow characteristics into the ramjet combustor. The results provide an engine inlet design capable of propelling the vehicle between the turbojet and scramjet phase of flight, allowing for the completion of its mission profile. Compromises in the design, however, had to be made in order to allow for optimisation of other propulsion systems including the scramjet nozzle and aerodynamics of the vehicle; it was concluded that these compromises were justified as the vehicle uses the ramjet engine for a minority of the flight profile as it transitions between low supersonic to hypersonic conditions.

show abstract

An MPI-CUDA approach for hypersonic flows with detailed state-to-state air kinetics using a GPU cluster

Cited by 51 publications

References 51 publications

On GPU Implementation of the Island Model Genetic Algorithm for Solving the Unequal Area Facility Layout Problem

On GPU Implementation of the Island Model Genetic Algorithm for Solving the Unequal Area Facility Layout Problem

A Multi‐GPU Parallel Algorithm in Hypersonic Flow Computations

Ramjet Compression System for a Hypersonic Air Transportation Vehicle Combined Cycle Engine

Contact Info

Product

Resources

About