## Hilbert R-tree: An improved R-tree using fractals (1994)

### Cached

### Download Links

- [www.cs.cmu.edu]
- [ftp.cs.umd.edu]
- CiteULike
- DBLP

### Other Repositories/Bibliography

Citations: | 185 - 9 self |

### BibTeX

@INPROCEEDINGS{Kamel94hilbertr-tree:,

author = {Ibrahim Kamel and Christos Faloutsos},

title = {Hilbert R-tree: An improved R-tree using fractals},

booktitle = {},

year = {1994},

pages = {500--509}

}

### Years of Citing Articles

### OpenURL

### Abstract

We propose a new R-tree structure that outperforms all the older ones. The heart of the idea is to facilitate the deferred splitting approach in R-trees. This is done by proposing an ordering on the R-tree nodes. This ordering has to be 'good', in the sense that it should group 'similar' data rectangles together, to minimize the area and perimeter of the resulting minimum bounding rectangles (MBRs). Following [19] we have chosen the so-called '2D-c' method, which sorts rectangles according to the Hilbert value of the center of the rectangles. Given the ordering, every node has a welldefined set of sibling nodes; thus, we can use deferred splitting. By adjusting the split policy, the Hilbert R-tree can achieve as high utilization as desired. To the contrary, the R -tree has no control over the space utilization, typically achieving up to 70%. We designed the manipulation algorithms in detail, and we did a full implementation of the Hilbert R-tree. Our experiments show that the '2-to-...

### Citations

2219 | R-Trees: A Dynamic Index Structure for Spatial Searching
- Guttman
- 1984
(Show Context)
Citation Context ...ectangle, retrieve all the elements that intersect it. A special case of the range query is the point query or stabbing query, where the query rectangle degenerates to a point. We focus on the R-tree =-=[15]-=- family of methods, which contains some of the most efficient methods that support range queries. The advantage of our method (and the rest of the R-treebased methods) over the methods that use linear... |

1174 |
The Design and Analysis of Spatial Data Structures
- Samet
- 1990
(Show Context)
Citation Context ...bert R-tree with other R-tree variants. Section 5 gives the conclusions and directions for future research. 2 Survey Several spatial access methods have been proposed. A recent survey can be found in =-=[26]-=-. These methods fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees [9] [1] or, equivalently,... |

1117 |
Multidimensional binary search trees used for associative senrching
- BENTLEY
- 1975
(Show Context)
Citation Context ...lity space [16, 8]; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees =-=[4]-=-, k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t.c.) One of the most promising approaches in the last class is the R-tree [15]: Compared to the transformation methods, R-trees work on the native... |

980 | The R*-tree: An efficient and robust access method for points and rectangles
- Beckmann, Kriegel, et al.
- 1990
(Show Context)
Citation Context ... of the Hilbert R-tree. Our experiments show that the '2-to-3' split policy provides a compromise between the insertion complexity and the search cost, giving up to 28% savings over the R \Gamma tree =-=[3]-=- on real data. 1 Introduction One of the requirements for the database management systems (DBMSs) of the near future is the ability to handle spatial data [28]. Spatial data arise in many applications... |

828 |
Computer Vision
- Ballard, Brown
- 1982
(Show Context)
Citation Context ...f the near future is the ability to handle spatial data [28]. Spatial data arise in many applications, including: Cartography [29]; Computer-Aided Design (CAD) [24] [14]; computer vision and robotics =-=[2]-=-; traditional databases, where a record with k attributes corresponds to a point in a k-d space; temporal This research was partially funded by the Institute for Systems Research (ISR), by the Nationa... |

380 |
The K-D-B-tree: a search structure for large multidimensional dynamic indexes
- Robinson
- 1981
(Show Context)
Citation Context ...]; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees =-=[25]-=-, hB-trees [21], cell-trees [13] e.t.c.) One of the most promising approaches in the last class is the R-tree [15]: Compared to the transformation methods, R-trees work on the native space, which has ... |

297 | C.: The r + -tree: a dynamic index for multi-dimensional objects
- Sellis, Roussopoulos, et al.
- 1987
(Show Context)
Citation Context ...ee, the quadratic split algorithm is the one that achieves the best trade-off between splitting time and search performance. Subsequent work on R-trees includes the work by Greene [11], the R + -tree =-=[27]-=-, R-trees using Minimum Bounding Polygons [17], and finally, the R -tree [3], which seems to have the best performance among the R-tree variants. The main idea in the R -tree is the concept of forced ... |

222 | On packing R-trees
- Kamel, Faloutsos
- 1993
(Show Context)
Citation Context ...This ordering has to be 'good', in the sense that it should group 'similar' data rectangles together, to minimize the area and perimeter of the resulting minimum bounding rectangles (MBRs). Following =-=[19]-=- we have chosen the so-called '2D-c' method, which sorts rectangles according to the Hilbert value of the center of the rectangles. Given the ordering, every node has a welldefined set of sibling node... |

198 |
The Fractal Geometry of Nature (W.H
- Mandelbrot
- 1982
(Show Context)
Citation Context ...ately rotated and/or reflected. Figure 1 also shows the Hilbert curves of order 2 and 3. When the order of the curve tends to infinity, the resulting curve is a fractal, with a fractal dimension of 2 =-=[22]-=-. The Hilbert curve can be generalized for higher dimensionalities. Algorithms to draw the two-dimensional curve of a given order, can be found in [12], [18]. An algorithm for higher dimensionalities ... |

185 |
Linear clustering of objects with multiple attributes
- Jagadish
- 1990
(Show Context)
Citation Context ...thods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] =-=[18]-=-; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t.c.) One of the most promising approaches in the last class is the R-tree [15]: C... |

181 | The hb-tree: a multiattribute indexing method with good guaranteed performance
- Lomet, Salzberg
- 1990
(Show Context)
Citation Context ... use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees =-=[21]-=-, cell-trees [13] e.t.c.) One of the most promising approaches in the last class is the R-tree [15]: Compared to the transformation methods, R-trees work on the native space, which has lower dimension... |

169 |
Spatial query processing in an object-oriented database system
- Orenstein
- 1986
(Show Context)
Citation Context ...fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering =-=[23]-=- or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t.c.) One of the most promising approaches ... |

140 | Fractals for Secondary Key Retrieval
- Faloutsos, Roseman
- 1989
(Show Context)
Citation Context ...: methods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves =-=[7]-=- [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t.c.) One of the most promising approaches in the last class is the R-tree [1... |

137 |
An effective way to represent quadtrees
- Gargantini
- 1982
(Show Context)
Citation Context ...rvey can be found in [26]. These methods fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees =-=[9]-=- [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t... |

107 |
Database systems: Achievements and opportunities
- SILBERSCHATZ, STONEBRAKER, et al.
- 1991
(Show Context)
Citation Context ... up to 28% savings over the R \Gamma tree [3] on real data. 1 Introduction One of the requirements for the database management systems (DBMSs) of the near future is the ability to handle spatial data =-=[28]-=-. Spatial data arise in many applications, including: Cartography [29]; Computer-Aided Design (CAD) [24] [14]; computer vision and robotics [2]; traditional databases, where a record with k attributes... |

97 |
The BANG File: a new Kind of Grid File
- Freeston
(Show Context)
Citation Context ...ess methods have been proposed. A recent survey can be found in [26]. These methods fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space =-=[16, 8]-=-; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [... |

63 |
An implementation and performance analysis of spatial data access methods
- Greene
- 1989
(Show Context)
Citation Context ...exity; among the three, the quadratic split algorithm is the one that achieves the best trade-off between splitting time and search performance. Subsequent work on R-trees includes the work by Greene =-=[11]-=-, the R + -tree [27], R-trees using Minimum Bounding Polygons [17], and finally, the R -tree [3], which seems to have the best performance among the R-tree variants. The main idea in the R -tree is th... |

60 |
Segment indexes: dynamic indexing techniques for multi-dimensional interval data
- Kolovson, Stonebraker
- 1991
(Show Context)
Citation Context ...ence Foundation under Grants IRI-9205273 and IRI-8958546 (PYI), with matching funds from EMPRESS Software Inc. and Thinking Machines Inc. databases, where time can be considered as one more dimension =-=[20]-=-; scientific databases with spatial-temporal data, such as the ones in the `Grand Challenge' applications [10], etc. In the above applications, one of the most typical queries is the range query: Give... |

58 |
Space-filling curves: Their generation and their application to bandwidth reduction
- Bially
- 1969
(Show Context)
Citation Context ...Hilbert curve can be generalized for higher dimensionalities. Algorithms to draw the two-dimensional curve of a given order, can be found in [12], [18]. An algorithm for higher dimensionalities is in =-=[5]-=-. The path of a space filling curve imposes a linear ordering on the grid points. Figure 1 shows 0 1 2 3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 H H H 1 2 3 Figure 1: Hilbert Curves of order 1, 2 and 3 ... |

55 |
Optimization Strategies for Spatial Query Processing
- Aref, Samet
- 1991
(Show Context)
Citation Context ... can be found in [26]. These methods fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space [16, 8]; methods that use linear quadtrees [9] =-=[1]-=- or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees [13] e.t.c.)... |

52 |
Spatial Search with Polyhedra
- Jagadish
- 1990
(Show Context)
Citation Context ...hat achieves the best trade-off between splitting time and search performance. Subsequent work on R-trees includes the work by Greene [11], the R + -tree [27], R-trees using Minimum Bounding Polygons =-=[17]-=-, and finally, the R -tree [3], which seems to have the best performance among the R-tree variants. The main idea in the R -tree is the concept of forced re-insert. When a node overflows, some of its ... |

42 |
Magic: A VLSI layout system
- Ousterhout, Hamachi, et al.
- 1984
(Show Context)
Citation Context ...he database management systems (DBMSs) of the near future is the ability to handle spatial data [28]. Spatial data arise in many applications, including: Cartography [29]; Computer-Aided Design (CAD) =-=[24]-=- [14]; computer vision and robotics [2]; traditional databases, where a record with k attributes corresponds to a point in a k-d space; temporal This research was partially funded by the Institute for... |

27 |
The grid file: a data structure to support proximity queries on spatial objects
- Hinrichs, Nievergelt
- 1983
(Show Context)
Citation Context ...ess methods have been proposed. A recent survey can be found in [26]. These methods fall in the following broad classes: methods that transform rectangles into points in a higher dimensionality space =-=[16, 8]-=-; methods that use linear quadtrees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [... |

23 |
Codes for Partial Match and Range Queries
- Gray
- 1988
(Show Context)
Citation Context ...visits all the points in a k-dimensional grid exactly once and never crosses itself. The Z-order (or Morton key order, or bit-interleaving, or Peano curve), the Hilbert curve, and the Gray-code curve =-=[6]-=- are examples of space filling curves. In [7], it was shown experimentally that the Hilbert curve achieves the best clustering among the three above methods. Next we provide a brief introduction to th... |

19 |
Challenges: High performance computing and communications. The FY
- Grand
- 1992
(Show Context)
Citation Context ... and Thinking Machines Inc. databases, where time can be considered as one more dimension [20]; scientific databases with spatial-temporal data, such as the ones in the `Grand Challenge' applications =-=[10]-=-, etc. In the above applications, one of the most typical queries is the range query: Given a rectangle, retrieve all the elements that intersect it. A special case of the range query is the point que... |

15 |
N-trees: large ordered indexes for multidimensional space. hesented at the Lincoln Institute of Land Policy's Colloquium on Spatial Mathematical Algorithms for Microcornputer Land Data Systems
- White
- 1983
(Show Context)
Citation Context ...tion One of the requirements for the database management systems (DBMSs) of the near future is the ability to handle spatial data [28]. Spatial data arise in many applications, including: Cartography =-=[29]-=-; Computer-Aided Design (CAD) [24] [14]; computer vision and robotics [2]; traditional databases, where a record with k attributes corresponds to a point in a k-d space; temporal This research was par... |

13 |
An Algorithm for Displaying a Class of Space-filling Curves
- Griffiths
- 1986
(Show Context)
Citation Context ...urve is a fractal, with a fractal dimension of 2 [22]. The Hilbert curve can be generalized for higher dimensionalities. Algorithms to draw the two-dimensional curve of a given order, can be found in =-=[12]-=-, [18]. An algorithm for higher dimensionalities is in [5]. The path of a space filling curve imposes a linear ordering on the grid points. Figure 1 shows 0 1 2 3 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15... |

12 |
New features for relational database systems to support CAD applications
- Guttman
- 1984
(Show Context)
Citation Context ...tabase management systems (DBMSs) of the near future is the ability to handle spatial data [28]. Spatial data arise in many applications, including: Cartography [29]; Computer-Aided Design (CAD) [24] =-=[14]-=-; computer vision and robotics [2]; traditional databases, where a record with k attributes corresponds to a point in a k-d space; temporal This research was partially funded by the Institute for Syst... |

9 |
The Cell Tree: An Index for Geometric Data,’’ Memorandum No
- Gunther
- 1986
(Show Context)
Citation Context ...rees [9] [1] or, equivalently, the z-ordering [23] or other space filling curves [7] [18]; and finally, methods based on trees (R-tree [15], k-d-trees [4], k-d-B-trees [25], hB-trees [21], cell-trees =-=[13]-=- e.t.c.) One of the most promising approaches in the last class is the R-tree [15]: Compared to the transformation methods, R-trees work on the native space, which has lower dimensionality; compared t... |