کد مقاله | کد نشریه | سال انتشار | مقاله انگلیسی | نسخه تمام متن |
---|---|---|---|---|
511061 | 865814 | 2014 | 9 صفحه PDF | دانلود رایگان |
• Meshless formulation suitable for parallel implementation is presented..
• The straightforward OpenMP parallelization is applied on the proposed solution procedure..
• Super linear speedup is identified and explained through the measurements of CPU counters.
• The effect of memory architecture on the speedup is discussed.
The performance of the parallel implementation of the local meshless numerical method in solving system of coupled partial differential equations is explored. Presented numerical approach makes the computation convenient for parallel implementation using OpenMP based parallelisation. The numerical experiments are performed on the de Vahl Davis natural convection case, with superlinear computational speedup regime identified. The phenomenon is further investigated through measurements of the central processing unit cache hit rates. It is demonstrated that the accumulation of L3 caches governs the superlinear speedup. Considering the presented analyses, basic rules for effective computation strategy regarding the multicore computations are suggested.
Journal: Computers & Structures - Volume 133, March 2014, Pages 30–38