One of the most essential and challenging components in climate modeling is the atmospheric model. To solve the multi-physical atmospheric equations, developers have to face extremely complex stencil kernels that are costly in terms of both computing and memory resources. This paper aims to accelerate the solution of the global shallow water equations (SWEs), which is one of the most essential equation sets that describes the atmospheric dynamics. We first design a hybrid methodology that

more »

... s both the host CPU cores and the FPGA accelerators to work in parallel. Through a careful adjustment of the computational domains, we achieve a balanced resource utilization and a further improvement of the overall performance. By decomposing the resource-demanding SWEs kernel, we manage to map the double-precision algorithm into 3 FPGAs. Moreover, by using fixed-point and reduced-precision floating point arithmetic, we manage to build a fully pipelined mixed-precision design on a single FPGA, which can perform 428 floating-point and 235 fixed-point operations per cycle. The mixed-precision design with 4 FPGAs running together can achieve a speed up of 20 over a fully-optimized design on a CPU rack with two 8-core processors, and is 8 times faster than the fully-optimized Kepler GPU design. As for the power efficiency, the mixed-precision design with 4 FPGAs is 10 times more power efficient than a Tianhe-1A supercomputer node.