Boost.Geometry R-tree - speeding up geographical computation.

Learn how to accelerate geographical computation with Boost Geometry's R-tree, a parameterized spatial index for storing and querying points, boxes, and segments, ideal for applications like routing, location-based services, and GIS.

Key takeaways

The Boost Geometry R-tree is a spatial index that can be used for geographic computation.
The R-tree is parameterized and can store various types, including points, boxes, and segments.
There are three balancing algorithms: linear, quadratic, and R-star tree versions.
The size of a node is defined by the page size, and the number of elements stored in a node affects the time spent on creating the R-tree.
The R-tree is self-balanced and can be created using different algorithms, including packing and bulk insertion.
The self-balancing algorithm tries to maintain a balance between the number of elements in each node.
The R-tree can be used for searching and KNN (k-nearest neighbors) searching, and it is used by MySQL for GIS features.
The Boost Geometry library also provides primitives, algorithms, and data structures for geographic computation.
The R-tree can be used with different data types, such as Cartesian points, boxes, or segments, and it supports the use of different spheroids and coordinates systems.
The Boost Geometry library is used by many organizations, including Oracle.
The R-tree is used for spatial searching and KNN searching, which can be used in many applications, such as routing, pathfinding, and location-based services.
The R-tree can be used with different query types, including range queries, nearest neighbor queries, and spatial joins.
The R-tree is scalable and can handle large datasets.
The R-tree can be used in many applications, including geographic information systems, computer-aided design, and video games.

Boost.Geometry R-tree - speeding up geographical computation.

More talks