Data parallel implementation methods of a stiff chemical non-equilibrium flow solver: parallelization, vectorization and I/O

2000 ◽  
Author(s):  
Shijun Diao ◽  
T. Fujiwara
Keyword(s):  
Parallel Implementation ◽  
Equilibrium Flow ◽  
Flow Solver ◽  
Data Parallel ◽  
Non Equilibrium

Analysis of numerical results in high temperature congealment and chemistry non-equilibrium flow field

2002 ◽  
Vol 1 (1) ◽  
pp. 26-34
Author(s):  
Hong-tao Zheng ◽  
Zhi-yong Tan ◽  
Hai-ou Sun ◽  
Chun-liang Zhou ◽  
Zhi-ming Li
Keyword(s):  
High Temperature ◽  
Flow Field ◽  
Numerical Results ◽  
Equilibrium Flow ◽  
Non Equilibrium

AN EVALUATION OF MULTIPLE FEED-FORWARD NETWORKS ON GPUs

2011 ◽  
Vol 21 (01) ◽  
pp. 31-47 ◽  
Author(s):  
NOEL LOPES ◽  
BERNARDETE RIBEIRO
Keyword(s):  
Graphics Processing Unit ◽  
Parallel Implementation ◽  
Low Cost ◽  
Back Propagation ◽  
General Purpose ◽  
Training System ◽  
Graphics Hardware ◽  
Processing Unit ◽  
Data Parallel ◽  
Graphics Processing

The Graphics Processing Unit (GPU) originally designed for rendering graphics and which is difficult to program for other tasks, has since evolved into a device suitable for general-purpose computations. As a result graphics hardware has become progressively more attractive yielding unprecedented performance at a relatively low cost. Thus, it is the ideal candidate to accelerate a wide variety of data parallel tasks in many fields such as in Machine Learning (ML). As problems become more and more demanding, parallel implementations of learning algorithms are crucial for a useful application. In particular, the implementation of Neural Networks (NNs) in GPUs can significantly reduce the long training times during the learning process. In this paper we present a GPU parallel implementation of the Back-Propagation (BP) and Multiple Back-Propagation (MBP) algorithms, and describe the GPU kernels needed for this task. The results obtained on well-known benchmarks show faster training times and improved performances as compared to the implementation in traditional hardware, due to maximized floating-point throughput and memory bandwidth. Moreover, a preliminary GPU based Autonomous Training System (ATS) is developed which aims at automatically finding high-quality NNs-based solutions for a given problem.

Effects of Chemical Reaction Models on Simulations of Scramjet Nozzle Flow

2014 ◽  
Vol 490-491 ◽  
pp. 931-935
Author(s):  
Xiao Yuan Zhang ◽  
Li Zi Qin ◽  
Yu Liu
Keyword(s):  
Kinetic Models ◽  
Chemical Kinetic ◽  
Numerical Results ◽  
Supersonic Combustion ◽  
Equilibrium Flow ◽  
Reaction Models ◽  
Total Temperature ◽  
Supersonic Combustion Ramjet ◽  
Quantity Of Heat ◽  
Non Equilibrium

The chemical non-equilibrium flow of supersonic combustion ramjet (scramjet) nozzle is numerical simulated with different chemical kinetic models to research the effects on numerical results of the nozzle performance. The numerical results show that total temperature is increased due to the recombination of dissociation compositions and the combustion of the residual fuel. The effect of the combustion of the residual fuel is more obvious in this paper, and the effect to the performance of the nozzle is noticeable. The species of the compositions in the models influence the quantity of heat sending out when it get equilibrium, so the 9-species chemical kinetic models are more suitable in the simulation of the scramjet nozzle chemical non-equilibrium flows.

Sign in / Sign up

Export Citation Format

Share Document