abstract = {The Crank-Nicolson (CN) method is a well-known time integrator for evolutionary partial differential equations (PDEs) arising in many real-world applications. Since the solution at any time depends on the solution at previous time steps, the CN method will be inherently difficult to parallelize. In this paper, we consider a parallel method for the solution of evolutionary PDEs with the CN scheme. Using an all-at-once approach, we can solve for all time steps simultaneously using a parallelizable over time preconditioner within a standard iterative method. Due to the diagonalization of the proposed preconditioner, we can prove that most eigenvalues of preconditioned matrices are equal to 1 and the others lie in the set: $\left\{z\in\mathbb{C}: 1/(1 + \alpha) < |z| < 1/(1 - \alpha)~{\rm and}~\Re{e}(z) > 0\right\}$, where $0 < \alpha < 1$ is a free parameter. Meanwhile, the efficient implementation of this proposed preconditioner is described and a mesh-independent convergence rate of the preconditioned GMRES method is derived under certain conditions. Finally, we will verify our theoretical findings via numerical experiments on financial option pricing partial differential equations.},
0 commit comments