Abstract
Parallel computational fluid dynamics is one of the key applications in the area of high-performance computations. The primary goals of this work are to develop a parallel unstructured finite-volume solver for all-speed flows based on a domain decomposition method, and to establish a comprehensive and intensive analysis method for the parallel performance. The numerical calculations of several typical flows using hundreds of CPU cores validate the accuracy and parallel performance of the parallel solver. The analysis of the decomposed efficiencies reveals the key factors that limit the parallel performance. This work is quite generic and can be extended to the large-scale parallel calculation and performance analysis of complex fluid flow and heat transfer problems.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
