Spatio-Temporal Data Management via Bit String Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing database systems face challenges in efficiently managing and searching vast amounts of spatio-temporal data due to extensive index update times, difficulties in multi-dimensional range searches, and imbalanced processing loads when using relational database management systems and distributed key-value stores.
Innovation Solution
An information management device that distributes spatio-temporal information and associated data across multiple key-value store nodes, converting data into one-dimensional bit strings for efficient storage and search, splitting these strings into upper and lower bits for balanced node distribution and reduced search loads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If spatio-temporal information is accumulated in an RDBMS with index updates, then data management capability is improved, but update time becomes excessively long when handling vast amounts of data
Solution Approach 1:
The patent segments the 64-bit spatio-temporal data into two 32-bit parts: upper 32 bits stored as key in distributed key-value stores, and lower 32 bits stored as value. This segmentation eliminates the need for traditional database indexes while maintaining efficient retrieval capability, resolving the contradiction between data management capability and update time.
2Adaptability or versatility
If secondary indexes are used in distributed key-value stores to store multiple values in one key, then search condition flexibility is improved, but processing load becomes imbalanced among nodes and performance deteriorates
Solution Approach 1:
The patent applies segmentation by dividing the 64-bit spatio-temporal data into upper and lower 32-bit parts, storing them separately as key and value in distributed key-value stores. This approach maintains search flexibility while distributing data evenly across nodes, avoiding the performance deterioration caused by imbalanced processing loads in secondary index approaches.
3Productivity
If spatio-temporal information is converted into a one-dimensional bit string for efficient storage, then data storage efficiency is improved, but multi-dimensional range search capability is lost
Solution Approach 1:
The patent segments the one-dimensional 64-bit spatio-temporal bit string into upper and lower 32-bit parts. The upper part (representing multi-dimensional spatial coordinates) is stored as key enabling efficient range searches, while the lower part (representing time information) is stored as value. This segmentation preserves both storage efficiency and multi-dimensional range search capability.
Solution Approach 2:
The patent transforms the one-dimensional bit string representation into a two-dimensional storage structure in distributed key-value stores, where the upper 32 bits form the key dimension and the lower 32 bits form the value dimension. This dimensional transformation enables efficient multi-dimensional range searches while maintaining storage efficiency.
Data Source
AI summary
An information management device includes a memory, and processing circuitry coupled to the memory and configured to convert said spatio-temporal information in storage object information into a one-dimensional bit string, split the converted one-dimensional bit string into an upper bit string and a lower bit string, and cause a storage target node to store at least the split upper bit string in a key and to store the split lower bit string and said associated data in a value of that key, and convert a range condition of spatio-temporal information of an object to be retrieved into one-dimensional bit string, split the converted one-dimensional bit string into an upper bit string and a lower bit string, retrieve a key from a search target node using at least the split upper bit string, and retrieve a value corresponding to the split lower bit string from values of the retrieved key.