System and method for data management
AssigneeKabushiki Kaisha Toshiba
InventorsMutsumi Fujihara, Shuichi Kamimura
Life of this patent
Estimating the term…
Abstract
The present invention provides a data management system that efficiently uses data areas and significantly increases search efficiency. According to this invention, in the data management system coordinate data 1 is stored in an index 2. An index generating means 3 generates the index from the coordinate data. A range search means 4 finds all coordinate data included in a specified range in the n-dimensional space, that is, one or more pieces of coordinate data which exist in an outmost area and which is included in a specified range. A nearest data search means 5 finds coordinate data nearest to a specified coordinate data. The index contains area nests in the form of code strings representing coordinates of dimensions and sequences of terms representing divisions.
Claims33
A data management system comprising:
- a storage part configured to store an index comprising n-dimensional coordinate data, the coordinate data being sorted into lexicographic order of area codes, each area code being associated with one of the n-dimensional coordinate data and generated by concatenating numeric substrings taken in a predetermined order from each coordinate of the associated one n-dimensional coordinate data, a length of each numeric substring being determined by a set of prefixes of division, one set for each of the n dimensions;
- a search part configured to search the index for specified coordinate data using the area codes;
- and
- an index changing part configured to perform one of adding a set of coordinates to said index and deleting a set of coordinates from said index, wherein the index changing part searches the index using said area code according to a binary search method and adds a new set of coordinates at a position consistent with the sort order, thereby adding the coordinate set to the index.
A data management method, comprising:
- sorting n-dimensional coordinate data stored in an index into lexicographic order of area codes, each area code being associated with one of the n-dimensional coordinate data and generated by concatenating numeric substrings taken in a predetermined order from each coordinate of the associated one n-dimensional coordinate data, a length of each numeric substring being determined by a set of prefixes of division, one set for each of the n dimensions;
- searching the index for specified coordinate data using the area codes;
- and
- performing one of adding a set of coordinates to said index and deleting a set of coordinates from said index, wherein the index is searched using said area code according to a binary search method and a new set of coordinates are added at a position consistent with the sort order, thereby adding the coordinate set to the index.
A computer-readable medium having stored therein a data management program, the program comprising:
- a means for causing a computer to sort n-dimensional coordinate data stored in an index into lexicographic order of area codes, each area code being associated with one of the n-dimensional coordinate data and generated by concatenating numeric substrings taken in a predetermined order from each coordinate of the associated one n-dimensional coordinate data, a length of each numeric substring being determined by a set of prefixes of division, one set for each of the n dimensions;
- a means for causing a computer to search the index for specified coordinate data using the area codes;
- and
- a means for causing a computer to perform one of adding the coordinates to said index and deleting the coordinates from said index, wherein the index is searched using said area code according to a binary search method and a new set of coordinates are added at a position consistent with the sort order, thereby adding the coordinate set to the index.
Dependencies are inferred from the claim text.