1: \begin{abstract}

2: We consider an efficient preconditioner for boundary integral equation

3: (BIE) formulations of the two-dimensional Stokes equations in porous

4: media.  While BIEs are well-suited for resolving the complex porous

5: geometry, they lead to a dense linear system of equations that is

6: computationally expensive to solve for large problems.  This expense is

7: further amplified when a significant number of iterations is required in

8: an iterative Krylov solver such as GMRES.  In this paper, we apply a

9: fast inexact direct solver, the inverse fast multipole method (IFMM), as

10: an efficient preconditioner for GMRES. This solver is based on the

11: framework of $\mH^{2}$-matrices and uses low-rank compressions to

12: approximate certain matrix blocks. It has a tunable accuracy

13: $\varepsilon$ and a computational cost that scales as $\mO (N \log^2

14: 1/\varepsilon)$. We discuss various numerical benchmarks that validate

15: the accuracy and confirm the efficiency of the proposed method. We

16: demonstrate with several types of boundary conditions that the

17: preconditioner is capable of significantly accelerating the convergence

18: of GMRES when compared to a simple block-diagonal preconditioner,

19: especially for pipe flow problems involving many pores.

20: \end{abstract}

21: