Spatial Search Index with Attribute Hint Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RDBMS systems face performance issues when performing spatial searches under combined spatial and attribute conditions, as they require separate processing and merging of results, leading to increased processing time.
Innovation Solution
A spatial search index is enhanced by incorporating attribute conditions, allowing for the storage of additional information as hint data, enabling fast searching by using a tree structure with location and attribute information, and employing bit arrays for efficient attribute filtering.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If separate processing is used for spatial search and attribute search, then each search can be performed independently, but processing time increases due to separate processing and merging of results
Solution Approach 1:
The patent merges spatial search and attribute search into a single integrated spatial search index. The index stores both spatial information (location, geometry) and attribute information (category, name, and hint data representing attribute conditions) together, allowing simultaneous search for both spatial and attribute conditions in one operation, thereby eliminating the need for separate processing and result merging.
Solution Approach 2:
The spatial search index is designed as a multi-functional structure that can handle both spatial queries and attribute queries. By incorporating hint data that represents attribute conditions directly into the index nodes, the system enables a single index structure to serve multiple search purposes, improving versatility while reducing processing time.
2Device complexity
If attribute information is stored separately from spatial information, then data organization is simpler, but search performance deteriorates due to additional processing steps
Solution Approach 1:
The patent combines spatial information and attribute information into a unified spatial search index structure. Each node in the tree contains both spatial data (bounding box, geometry) and attribute data (category, name, hint), enabling simultaneous indexing and search of both types of information without requiring separate data structures or additional processing steps.
3Adaptability or versatility
If conventional RDBMS is used for spatial search, then standard database functions are available, but performance is poor for combined spatial and attribute conditions
Solution Approach 1:
The patent performs preliminary action by pre-processing and storing attribute conditions as hint data during index construction. This preliminary organization of attribute information within the index structure enables fast search operations without requiring complex runtime processing, achieving high performance for combined spatial and attribute conditions.
Data Source
AI summary
A method for processing data performed in a data processing system having a storage unit is provided which includes: creating a spatial index comprising a leaf containing location information and attribute information with respect to a plurality of objects to be searched, and storing the spatial index in the storage unit; and referring to the spatial index stored in the storage unit in response to an input of a search request including location information and attribute information for searching the object to be searched, and extracting the object to be searched that agrees with the search request. With this configuration, a spatial search with subsidiary condition search can be processed at high speed, without merging results of both searches.