Multidimensional Access Methods: Important Factor for Current and Next Decade’s Applications in Spatial Databases

Abstract: In spatial databases search operations take an important role. These operations consist of the point query (find all objects that contain a given search point), the range query (find all objects that overlap a given search range) [GAE98] and the nearest neighbor query (find k objects (k> = 1) that are closest to a given object). They are very costly operations. Their performance is affected by both CPU-time and IO-cost. The long history of researches on spatial databases has resulted in many multidimensional access methods to efficiently support such operations. Each of them has strengths and deficiencies as well. This paper preliminary develops taxonomy for these multidimensional access methods and describes some prominent multidimensional access methods, which have been recently introduced as well as their comparative studies. Moreover, the important role of multidimensional access methods for supporting real-world applications in the next decade is also discussed