On The Performance of Prefix-Sum Parallel Kalman Filters and Smoothers on GPUs

TL;DR

This paper evaluates the performance of parallel prefix-sum algorithms for Kalman filters and smoothers on GPUs, introduces a novel parallel two-filter smoother, and provides open-source Julia implementations.

Contribution

It provides a comprehensive experimental analysis of all-prefix-sum algorithms for Kalman filtering on GPUs and proposes a new parallel two-filter smoother.

Findings

Parallel scan algorithms' efficiency varies with implementation and hardware.

The novel parallel two-filter smoother shows promising performance improvements.

Open-source Julia code facilitates reproducibility and further research.

Abstract

This paper presents an experimental evaluation of parallel-in-time Kalman filters and smoothers using graphics processing units (GPUs). In particular, the paper evaluates different all-prefix-sum algorithms, that is, parallel scan algorithms for temporal parallelization of Kalman filters and smoothers in two ways: by calculating the required number of operations via simulation, and by measuring the actual run time of the algorithms on real GPU hardware. In addition, a novel parallel-in-time two-filter smoother is proposed and experimentally evaluated. Julia code for Metal and CUDA implementations of all the algorithms is made publicly available.