The Case for Learned Spatial Indexes

Abstract

Spatial data is ubiquitous. Massive amounts of data are generated every day from billions of GPS-enabled devices such as cell phones, cars, sensors, and various consumerbased applications such as Uber, Tinder, location-tagged posts in Facebook, Twitter, Instagram, etc. This exponential growth in spatial data has led the research community to focus on building systems and applications that can process spatial data efficiently. In the meantime, recent research has introduced learned index structures. In this work, we use techniques proposed from a state-of-the art learned multi-dimensional index structure (namely, Flood) and apply them to five classical multi-dimensional indexes to be able to answer spatial range queries. By tuning each partitioning technique for optimal performance, we show that (i) machine learned search within a partition is faster by 11.79% to 39.51% than binary search when using filtering on one dimension, (ii) the bottleneck for tree structures is index lookup, which could potentially be improved by linearizing the indexed partitions (iii) filtering on one dimension and refining using machine learned indexes is 1.23x to 1.83x times faster than closest competitor which filters on two dimensions, and (iv) learned indexes can have a significant impact on the performance of low selectivity queries while being less effective under higher selectivities.

The Case for Distance-Bounded Spatial Approximations

“…A common approach is to map 2D cells into a 1D domain by enumerating them with a space-filling curve such as the Hilbert or Z curve. As we will show, we can achieve much higher lookup performance with linearized cells, even compared to well-tuned 2D spatial indexes [15]. Polygon Indexing.…”

“…In our experiment, we use 39,200 polygons corresponding to the NYC Census regions (query polygons) and 1.2B points from the NYC taxi data set (years 2009 to 2016) [20]. We implemented the kd-tree, Quadtree, and STR-packed R-tree baselines based on recent research [15]. For the Boost R * -tree, we chose the bulk-loading Figure 4(a) shows the cumulative query time to find the total number of points inside the query polygons, while varying the precision of the raster approximation (i.e., number of approximating cells per query polygon).…”