A Bit-Compatible Shared Memory Parallelization for ILU(k) Preconditioning and a Bit-Compatible Generalization to Distributed Memory

TL;DR

This paper introduces TPILU(k), a parallel ILU(k) preconditioner that maintains stability and bit-compatibility with sequential ILU(k), significantly improving performance on multi-core and distributed systems.

Contribution

The paper presents the first efficient parallel ILU(k) preconditioner that preserves stability and bit-compatibility, with optimizations for multi-core and distributed computing environments.

Findings

TPILU(k) runs faster with more cores while maintaining stability.

Achieves up to 9x speedup on 16-core systems.

Attains 50x speedup on 80-node clusters for large sparse matrices.

Abstract

ILU(k) is a commonly used preconditioner for iterative linear solvers for sparse, non-symmetric systems. It is often preferred for the sake of its stability. We present TPILU(k), the first efficiently parallelized ILU(k) preconditioner that maintains this important stability property. Even better, TPILU(k) preconditioning produces an answer that is bit-compatible with the sequential ILU(k) preconditioning. In terms of performance, the TPILU(k) preconditioning is shown to run faster whenever more cores are made available to it --- while continuing to be as stable as sequential ILU(k). This is in contrast to some competing methods that may become unstable if the degree of thread parallelism is raised too far. Where Block Jacobi ILU(k) fails in an application, it can be replaced by TPILU(k) in order to maintain good performance, while also achieving full stability. As a further optimization, TPILU(k) offers an optional level-based incomplete inverse method as a fast approximation for the original ILU(k) preconditioned matrix. Although this enhancement is not bit-compatible with classical ILU(k), it is bit-compatible with the output from the single-threaded version of the same algorithm. In experiments on a 16-core computer, the enhanced TPILU(k)-based iterative linear solver performed up to 9 times faster. As we approach an era of many-core computing, the ability to efficiently take advantage of many cores will become ever more important. TPILU(k) also demonstrates good performance on cluster or Grid. For example, the new algorithm achieves 50 times speedup with 80 nodes for general sparse matrices of dimension 160,000 that are diagonally dominant.