1: \begin{abstract}

2: Algebraic multigrid (AMG) solvers and preconditioners are some of the fastest numerical methods to solve linear systems,

3: particularly in a parallel environment, scaling to hundreds of thousands of cores. Most AMG methods and theory

4: assume a symmetric positive definite operator. This paper presents a new variation on classical AMG for

5: nonsymmetric matrices (denoted $\ell$AIR), based on a local approximation to the ideal restriction operator, coupled with F-relaxation.

6: A new block decomposition of the AMG error-propagation operator is used for a spectral analysis of convergence, and the efficacy

7: of the algorithm is demonstrated on systems arising from the discrete form of the advection-diffusion-reaction equation. $\ell$AIR is

8: shown to be a robust solver for various discretizations of the advection-diffusion-reaction equation, including time-dependent and

9: steady-state, from purely advective to purely diffusive. Convergence is robust for discretizations on unstructured meshes and using

10: higher-order finite elements, and is particularly effective on upwind discontinuous Galerkin discretizations. Although the implementation used

11: here is not parallel, each part of the algorithm is highly parallelizable, avoiding common multigrid adjustments for strong advection such as

12: line-relaxation and K- or W-cycles that can be effective in serial, but suffer from high communication costs in parallel, limiting their scalability.

13: \end{abstract}

14: