Browse

Scientific codes are usually highly parallelised and executed on heterogeneous architectures. Nowadays, it is common to use graphics accelerators (GPUs) to speed up data-parallel algorithms, and in the meantime, the already existing CPUs can help in this work. Distributing the jobs between systems is always a difficult problem, especially when the processing units have different runtime environments and hardware architectures. There are several attempts for static and dynamic load balancing, but most of these are ...

This paper presents a novel heat transfer model for the estimation of heat transfer coefficients without any information about the thermal boundary conditions. The massively parallel implementation (using the latest CUDA 8.0 framework and NVIDIA graphics cards) of the model gives the opportunity to use multiple graphics accelerators to execute the heat transfer simulations. This leads to very impressive performance which makes it possible to run thousands of simulations within a reasonable time. There are several ...