Search

For computational fluid dynamics (CFD) applications with a large number of grid points/cells, parallel computing is a common efficient strategy to reduce the computational time. How to achieve the best performance in the modern supercomputer system, especially with heterogeneous computing resources such as hybrid CPU+GPU, or a CPU + Intel Xeon Phi (MIC) co-processors, is still a great challenge.An in-house parallel CFD code capable of simulating three dimensional structured grid applications is developed and tested in this study. Several methods of parallelization, performance optimization and code tuning both in the CPU-only homogeneous system and in the heterogeneous system are proposed based on identifying potential parallelism of applications, balancing the work load among all kinds of computing devices, tuning the multi-thread code toward better performance in intra-machine node with hundreds of CPU/MIC cores, and optimizing the communication among inter-nodes, inter-cores, and between CPUs and MICs.Some benchmark cases from model and/or industrial CFD applications are tested on the Tianhe-1A and Tianhe-2 supercomputer to evaluate the performance. Among these CFD cases, the maximum number of grid cells reached 780 billion. The tuned solver successfully scales up to half of the entire Tianhe-2 supercomputer system with over 1.376 million of heterogeneous cores. The test results and performance analysis are discussed in detail. [ABSTRACT FROM AUTHOR]

To improve the numerical simulation accuracy of predicting the cavitating flow in centrifugal pumps, a new developed cavitating flow computational model, named two-phase three-component computational model, is proposed. To evaluate its performance, the comparisons are carried out among CFX default cavitating flow computational model, the optimal computational model and the proposed model, concerning a centrifugal pump. Meanwhile, the experiments are conducted to validate the calculations, including pump head drop test and cavitating flow visualization at the pump inlet. The results show that the two-phase three-component computational model has better performance both on predicting pump head drop and bubble structure. It is observed in the current work that the sheet cavity firstly occurs on the leading edge of the suction side of the impeller. With the cavitation number decreasing, the cavity length grows rapidly along the suction side towards impeller outlet. Simultaneously, the cavity starts to extend to the pressure side. Based on the analysis of the blade pressure load on the blade surface and total pressure coefficient distribution in the channel, it is found that the cavitation has great effect on blade pressure load. Besides, the total pressure coefficient drops mainly take place on the upstream, that is to say, the growth of cavitation affects downstream rarely. [ABSTRACT FROM AUTHOR]