Set intersection algorithms between sorted lists are important in triangles counting, community d... more Set intersection algorithms between sorted lists are important in triangles counting, community detection in graph analysis and in search engines where the intersection is computed between queries and inverted indexes. Many researches use GPU techniques for solving this intersection problem. The majority of these techniques focus on improving the level of parallelism by reducing redundant comparisons and distributing the workload among GPU threads. In this paper, we propose the GPU Test with Jumps (GTWJ) algorithm to compute the intersection between sorted lists using a new data structure. The idea of GTWJ is to group the data, of each sorted list, into a set of sequences. A sequence is identified by a key and is handled by a thread. Intersection is computed between sequences with the same key. This key allows skipping data packets in parallel if the keys do not match. A counter is used to avoid useless tests between cells of sequences with different lengths. Experiments on the data...
Résumé. L’opérateur skyline est devenu un paradigme dans les bases de données. Il consiste à loca... more Résumé. L’opérateur skyline est devenu un paradigme dans les bases de données. Il consiste à localiser Sky l’ensemble des points d’un espace vectoriel qui ne sont pas dominés. Cet opérateur est utile lorsqu’on n’arrive pas à se décider dans les situations conflictuelles. Le calcul des requêtes skyline est pénalisé par le nombre de points que peuvent contenir les bases de données. Dans ce papier, nous présentons une solution analytique pour la réduction de l’espace candidat et nous proposons une méthode efficace pour le calcul de ce type de requêtes
Skyline queries are important in many fields, especially for decision making. In this context, ob... more Skyline queries are important in many fields, especially for decision making. In this context, objects or tuples of databases are defined according to some numerical and non numerical attributes. The skyline operator acts on the numerical ones. The algorithms that implements this skyline operator are genrally of progressive or non progressive. The progressive ones return the skyline operator during its execution while non preogressive alogrithms return the result at the end of its execution. This paper presents a new progressive algorithm for computing the skyline points. This algorithm is based on sorting as a preprocessing of the input. The authors present new theorems for deducing promptly the first skyline points and reducing the candidate space. A new version of Divide-and-Conquer algorithm is used for computing the final skyline. Intensive experimentations on both real and synthetic datasets show that our algorithm presents best performance comparatively to other methods.
Set intersection algorithms between sorted lists are important in triangles counting, community d... more Set intersection algorithms between sorted lists are important in triangles counting, community detection in graph analysis and in search engines where the intersection is computed between queries and inverted indexes. Many researches use GPU techniques for solving this intersection problem. The majority of these techniques focus on improving the level of parallelism by reducing redundant comparisons and distributing the workload among GPU threads. In this paper, we propose the GPU Test with Jumps (GTWJ) algorithm to compute the intersection between sorted lists using a new data structure. The idea of GTWJ is to group the data, of each sorted list, into a set of sequences. A sequence is identified by a key and is handled by a thread. Intersection is computed between sequences with the same key. This key allows skipping data packets in parallel if the keys do not match. A counter is used to avoid useless tests between cells of sequences with different lengths. Experiments on the data...
Résumé. L’opérateur skyline est devenu un paradigme dans les bases de données. Il consiste à loca... more Résumé. L’opérateur skyline est devenu un paradigme dans les bases de données. Il consiste à localiser Sky l’ensemble des points d’un espace vectoriel qui ne sont pas dominés. Cet opérateur est utile lorsqu’on n’arrive pas à se décider dans les situations conflictuelles. Le calcul des requêtes skyline est pénalisé par le nombre de points que peuvent contenir les bases de données. Dans ce papier, nous présentons une solution analytique pour la réduction de l’espace candidat et nous proposons une méthode efficace pour le calcul de ce type de requêtes
Skyline queries are important in many fields, especially for decision making. In this context, ob... more Skyline queries are important in many fields, especially for decision making. In this context, objects or tuples of databases are defined according to some numerical and non numerical attributes. The skyline operator acts on the numerical ones. The algorithms that implements this skyline operator are genrally of progressive or non progressive. The progressive ones return the skyline operator during its execution while non preogressive alogrithms return the result at the end of its execution. This paper presents a new progressive algorithm for computing the skyline points. This algorithm is based on sorting as a preprocessing of the input. The authors present new theorems for deducing promptly the first skyline points and reducing the candidate space. A new version of Divide-and-Conquer algorithm is used for computing the final skyline. Intensive experimentations on both real and synthetic datasets show that our algorithm presents best performance comparatively to other methods.
Uploads
Papers by Lougmiri Zekri