Telescope Columns for Spatial Data Indexing

Solution Overview

Problem

Current column-store technologies are inefficient in handling spatial data due to the overhead of tree-based index structures, which are not optimized for in-memory database systems and incur significant computational costs when processing geometric relationships.

Innovation Solution

The implementation of dimensional dictionaries for decomposing linestrings into line segments and complex shapes into simplexes, allowing for efficient compression and rapid scanning of spatial data to identify intersections using index clusters that approximate coordinate values, thereby reducing computational complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If tree-based index structures (R-Tree, K-d Tree, Quadtree) are used to add spatial features to column stores, then spatial query capability is improved, but device complexity and computational overhead increase significantly

Engineering Contradiction:
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent extracts the spatial indexing function from traditional tree-based structures and implements it through telescope columns that store coordinate data in a flattened, column-oriented format. This removes the complex tree node structures and linked lists while preserving spatial query capabilities through coordinate-based filtering and range queries on the extracted coordinate columns.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent segments spatial geometries into their constituent coordinate points and stores them in separate telescope columns (e.g., X coordinates in one column, Y coordinates in another). This segmentation eliminates the need for hierarchical tree structures while enabling efficient spatial queries through columnar operations on the segmented coordinate data.

Inventive Principle:
Principle #1Segmentation

2Ease of operation

If tree-based index structures are used for spatial data, then spatial access is enabled, but processing speed decreases due to linked node traversal

Engineering Contradiction:
Core Design Contradiction:
Ease of operationVSSpeed

Solution Approach 1:

The patent replaces the mechanical linked-node traversal of tree structures with direct memory access to columnar coordinate data. Instead of following pointers through hierarchical nodes, the system performs vectorized operations on contiguous coordinate arrays, eliminating pointer chasing and enabling cache-efficient processing.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent employs batched, periodic processing of coordinate data through telescope columns, where operations are applied systematically across coordinate arrays in regular patterns. This periodic processing of segmented coordinates replaces the irregular, variable-time traversal of tree nodes with predictable, optimized columnar operations.

Inventive Principle:
Principle #19Periodic action

3Ease of manufacture

If standard columns with floating-point numbers are used to store geographic coordinates, then implementation simplicity is improved, but spatial query efficiency deteriorates

Engineering Contradiction:
Core Design Contradiction:
Ease of manufactureVSProductivity

Solution Approach 1:

The patent introduces telescope columns as a dynamic data structure that extends standard columns with spatial-aware operations. While maintaining the simple columnar storage format, telescope columns add dynamic capabilities for spatial queries, geometric operations, and coordinate transformations, bridging the gap between simplicity and efficiency.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent makes telescope columns universal by enabling them to perform both standard columnar operations and specialized spatial queries. The same telescope column structure supports general-purpose data access while simultaneously providing optimized spatial operations like distance calculations, spatial joins, and geometric predicates without requiring separate index structures.

Inventive Principle:
Principle #6Universality (Multi-functionality)

4Productivity

If column stores use optimized scans of vector data, then in-memory processing performance is improved, but compatibility with traditional spatial index structures decreases

Engineering Contradiction:
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent transitions from the hierarchical dimension of tree structures to the columnar dimension of telescope columns. By organizing spatial data in vertical columns rather than hierarchical nodes, the system achieves compatibility with column-store optimized scans while maintaining spatial query capabilities through coordinate-based operations on the columnar data.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10642807B2Column store optimization using telescope columns
Publication Date: 2020.05.05
  • US10642807B2 patent drawing
  • US10642807B2 patent drawing
  • US10642807B2 patent drawing

AI summary

A data set of spatial data having a plurality of dimensions and including linestrings can be processing by decomposing each linestring of the plurality of linestrings into a plurality of line segments. Each coordinate dimension appears in at least one line segment of the plurality of line segments can be listed in one of a plurality of dimensional dictionaries that each correspond to a dimension of the plurality of dimensions. A linestring of the plurality of linestrings can be represented as a set of the line segments using the plurality of dimensional dictionaries.