Efficient Wind Speed Nowcasting with GPU-Accelerated Nearest Neighbors Algorithm

TL;DR

This paper introduces a GPU-accelerated nearest neighbors algorithm for wind nowcasting that efficiently processes large datasets, reconstructs wind fields accurately, and significantly speeds up computations compared to traditional methods.

Contribution

The paper presents a novel tensor-based algorithm optimized for GPU computing that reduces time and memory costs in wind data analysis, outperforming existing k-NN and data selection techniques.

Findings

Achieves several-fold speedup over traditional k-NN methods

Provides accurate wind field reconstruction from large datasets

Offers an open-source implementation and dataset for replication

Abstract

This paper proposes a simple yet efficient high-altitude wind nowcasting pipeline. It processes efficiently a vast amount of live data recorded by airplanes over the whole airspace and reconstructs the wind field with good accuracy. It creates a unique context for each point in the dataset and then extrapolates from it. As creating such context is computationally intensive, this paper proposes a novel algorithm that reduces the time and memory cost by efficiently fetching nearest neighbors in a data set whose elements are organized along smooth trajectories that can be approximated with piece-wise linear structures. We introduce an efficient and exact strategy implemented through algebraic tensorial operations, which is well-suited to modern GPU-based computing infrastructure. This method employs a scalable Euclidean metric and allows masking data points along one dimension. When applied, this method is more efficient than plain Euclidean k-NN and other well-known data selection methods such as KDTrees and provides a several-fold speedup. We provide an implementation in PyTorch and a novel data set to allow the replication of empirical results.