一种基于PETSc的热传导方程大规模并行求解策略

Parallel-computing Strategy for Large-scale Heat Equation Based on PETSc

提出了一种大规模热传导方程并行求解的策略,采用了分布式内存和压缩矩阵技术解决超大规模稀疏矩阵的存储及其计算,整合了多种Krylov子空间方法和预条件子技术来并行求解大规模线性方程组,基于面向对象设计实现了具体应用与算法的低耦合。在Linux机群系统上进行了性能测试,程序具有良好的加速比和计算性能。

A parallel-computing strategy was presented to solve the large-scale heat equations. The distributed memory and compressed matrices technology was adopted for both the process of storage and evaluation of large-scale sparse matrices. All kinds of Krylov subspaee methods and preconditioners were introduced to assemble and solve the linear systems of equations. The code implementation of this strategy was written in high-level abstractions based on object-oriented technology which promotes code reuse, flexibility and helps to deeouple issues of parallelism from algorithm choices. The experiments carried on Linux clusters demonstrate that this strategy has achieved desirable speedup and efficiency.