Document Database Time-Series Indexing for Spatiotemporal Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing time series databases face inefficiencies in storing and querying large volumes of time series data due to high indexing, ingestion, and processing costs, particularly when handling spatial and temporal constraints.
Innovation Solution
A system that stores time series data in a plurality of documents, indexing them based on time and spatial ranges using geohashes, allowing efficient querying and filtering by both temporal and spatial constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If time series data is stored in traditional databases with individual indexing for each data point, then data retrieval can be performed, but indexing, ingestion, and processing costs become excessively high for large volumes of data
Solution Approach 1:
The patent segments the time series data into multiple documents based on time ranges, with each document containing a subset of data points. This segmentation allows the system to index only the necessary documents rather than every individual data point, significantly reducing indexing overhead and improving query efficiency for large volumes of time series data
Solution Approach 2:
The patent introduces a new dimensional organization by creating time range metadata that spans across multiple data points. Instead of indexing each data point in isolation, the system creates an additional indexing dimension at the document level using time ranges, enabling efficient retrieval without indexing every individual point
2Speed
If all data points are indexed to enable fast querying, then query speed improves, but storage and processing costs increase significantly
Solution Approach 1:
The patent extracts only the essential time range metadata from each document and stores it separately for indexing purposes. This extraction allows the system to maintain fast query capabilities by indexing only the necessary metadata rather than duplicating or heavily indexing the actual data points, reducing storage and processing requirements
Solution Approach 2:
The patent creates lightweight copies of time range metadata (start time and end time) for each document and indexes these copies instead of the full data points. This copying approach enables efficient querying while consuming minimal storage and processing resources, as the index operates on compact metadata representations rather than complete data structures
3Adaptability or versatility
If the system supports complex spatiotemporal queries with both spatial and temporal constraints, then query capability improves, but system complexity increases
Solution Approach 1:
The patent merges spatial and temporal indexing into a unified document-level metadata structure. Each document contains both spatial range metadata and time range metadata that work together, allowing the system to handle complex spatiotemporal queries through a single integrated indexing mechanism rather than separate complex systems for each dimension
Data Source
AI summary
A system manages time series data stored in association with documents. The system stores a time series in multiple documents. The time series data includes a series of data points indexed in time order, and each document stores a subset of the series of data points. Each document stores metadata about a time range of the subset of series of data points. The system indexes the documents based on their time ranges. The system may be used for storing and querying spatiotemporal data. Accordingly, each document further stores spatial data and metadata representing a spatial range of geohashes. Responsive to receiving a query for one or more data points, the system selects one or more documents from the multiple documents that contain data points in the particular spatial and time range based on the indexes. The system performs the query on the data points.