No abstract

No abstract

CPU/GPU computing allows scientists to tremendously accelerate their numerical codes. In this paper, we port and optimize a double precision alternating direction implicit (ADI) solver for three-dimensional compressible Navier-Stokes equations from our in-house Computational Fluid Dynamics (CFD) software on heterogeneous platform. First, we implement a full GPU version of the ADI solver to remove a lot of redundant data transfers between CPU and GPU, and then design two fine-grain schemes, namely "one-thread-one-point" and "one-thread-one-line", to maximize the performance. Second, we present a dual-level parallelization scheme using the CPU/GPU collaborative model to exploit the computational resources of both multi-core CPUs and many-core GPUs within the heterogeneous platform. Finally, considering the fact that memory on a single node becomes inadequate when the simulation size grows, we present a trilevel hybrid programming pattern MPI-OpenMP-CUDA that merges fine-grain parallelism using OpenMP and CUDA threads with coarse-grain parallelism using MPI for inter-node communication. We also propose a strategy to overlap the computation with communication using the advanced features of CUDA and MPI programming. We obtain speedups of 6.0 for the ADI solver on one Tesla M2050 GPU in contrast to two Xeon X5670 CPUs. Scalability tests show that our implementation can offer significant performance improvement on heterogeneous platform.

As a conservative, high-order accurate, shock-capturing method, weighted essentially non-oscillatory (WENO) scheme have been widely used to effectively resolve complicated flow structures in computational fluid dynamics (CFD) simulations. However, using a high-order WENO scheme can be highly time-consuming, which greatly limits the CFD application's performance efficiency. In this paper, we present various parallel strategies base on the latest many-core platform such as NVIDIA Fermi GPU, NVIDIA Kepler GPU and Intel MIC coprocessor to accelerate a high-order WENO scheme. Comparison analysis of the two generations GPUs between Fermi and Kepler, and cross-platform performance analysis (focusing on Kepler GPU and MIC) are also detailed discussed. The experiments show that the Kepler GPU offers a clear advantage in contrast to the previous Fermi GPU maintaining exactly the same source code. Furthermore, while Kepler GPU can be several times faster than MIC without utilizing the increasingly available SIMD computing power on Vector Processing Unit (VPU), MIC can provide the computing capability equivalent to Kepler GPU when VPU is utilized. Our implementations and optimization techniques can serve as case studies for paralleling high-order schemes on many-core architectures.

No abstract

No abstract