Compression scheme for improving cache behavior in database systems
A scheme for accessing an index structure using a reference minimum bounding shape is disclosed. In one example embodiment, a reference minimum bounding shape that encloses two or more minimum bounding shapes may be identified from an index structure stored in memory. Each of the two or more minimum bounding shapes may correspond to a data object associated with a corresponding leaf node of the index structure. In one example embodiment, the index structure may be accessed using the reference minimum bounding shape. In one example embodiment, at least one minimum bounding shape of the two or more minimum bounding shapes may be represented in a relative representation calculated relative to the reference minimum bounding shape. Also disclosed are a method, a system and a non-transitory computer-readable storage medium for accomplishing the same scheme as described above.
1. A computer-implemented method, comprising:
identifying, from an index structure stored in memory, a reference minimum bounding shape that encloses two or more minimum bounding shapes, each of the two or more minimum bounding shapes corresponding to a data object associated with a corresponding leaf node of the index structure; and
compressing the index structure using the reference minimum bounding shape, the compressing including representing at least one minimum bounding shape of the two or more minimum bounding shapes in a relative representation calculated relative to the reference minimum bounding shape, the representing including associating coordinates of a corresponding point of one or more points of the at least one minimum bounding shape with a first value comprising a first number of significant bits fewer than a second number of significant bits representing a second value associated with absolute coordinates of the corresponding point, the first value being calculated relative to coordinates of a reference point of the reference minimum bounding shape.
2. The computer-implemented method of claim 1 , wherein the representing further comprises storing the relative representation of the at least one minimum bounding shape in a corresponding node of the index structure.
3. The computer-implemented method of claim 1 , further comprising:
compressing the relative representation of the at least one minimum bounding shape into a first quantized representation using a finite level of quantization.
4. The computer-implemented method of claim 3 , wherein the compressing comprises choosing the finite level of quantization from a given set of quantization levels.
5. The computer-implemented method of claim 3 , further comprising transforming a search query into a second quantized representation.
6. The computer-implemented method of claim 5 , further comprising comparing the first quantized representation with the second quantized representation.
7. The computer-implemented method of claim 3 , wherein the compressing further comprises storing the first quantized representation in a corresponding node of the index structure.
8. The computer-implemented method of claim 1 , wherein a first entry of a non-leaf node of the index tree has a corresponding quantized representation and a pointer to a corresponding child node while other entries of the non-leaf node have only corresponding quantized representations.
9. The computer-implemented method of claim 1 , wherein the reference minimal bounding shape is stored only in a root node of the index structure.
10. A computer-implemented system, comprising:
a memory storing an index structure; and
a processor operatively coupled to the memory, the processor configured to:
identify, from the index structure, a reference minimum bounding shape that encloses two or more minimum bounding shapes, each of the two or more minimum bounding shapes corresponding to a data object associated with a corresponding leaf node of the index structure; and
compress the index structure using the reference minimum bounding shape, the compressing including representing at least one minimum bounding shape of the two or more minimum bounding shapes in a relative representation calculated relative to the reference minimum bounding shape, the representing including associating coordinates of a corresponding point of one or more points of the at least one minimum bounding shape with a first value comprising a first number of significant bits fewer than a second number of significant bits representing a second value associated with absolute coordinates of the corresponding point, the first value being calculated relative to coordinates of a reference point of the reference minimum bounding shape.
11. The computer-implemented system of claim 10 , wherein the processor is further configured to store the relative representation of the at least one minimum bounding shape in a corresponding node of the index structure.
12. The computer-implemented system of claim 10 , wherein the processor is further configured to compress the relative representation of the at least one minimum bounding shape into a first quantized representation using a finite level of quantization.
13. The computer-implemented system of claim 12 , wherein the processor is further configured to choose the finite level of quantization from a given set of quantization levels.
14. The computer-implemented system of claim 12 , wherein the processor is further configured to transform a search query into a second quantized representation.
15. The computer-implemented system of 14 , wherein the processor is further configured to compare the first quantized representation with the second quantized representation.
16. The computer-implemented system of claim 12 , wherein the processor is further configured to store the first quantized representation in a corresponding node of the index structure.
17. The computer-implemented system of claim 10 , wherein the index structure is chosen from a plurality of tree structures including an R-tree, an R*-tree, an R+-tree and a Hilbert R-tree.
18. A non-transitory computer-readable storage medium storing instructions that, when executed by a computer, cause the computer to perform operations comprising:
identifying, from an index structure stored in memory, a reference minimum bounding shape that encloses two or more minimum bounding shapes, each of the two or more minimum bounding shapes corresponding to a data object associated with a corresponding leaf node of the index structure; and
compressing the index structure using the reference minimum bounding shape, the compressing including representing at least one minimum bounding shape of the two or more minimum bounding shapes in a relative representation calculated relative to the reference minimum bounding shape, the representing including associating coordinates of a corresponding point of one or more points of the at least one minimum bounding shape with a first value comprising a first number of significant bits fewer than a second number of significant bits representing a second value associated with absolute coordinates of the corresponding point, the first value being calculated relative to coordinates of a reference point of the reference minimum bounding shape.