An application of number theory to the organization of raster-graphics memory
Journal of the ACM (JACM) - The MIT Press scientific computation series
Optimal file distribution for partial match retrieval
SIGMOD '88 Proceedings of the 1988 ACM SIGMOD international conference on Management of data
Optimal disk allocation for partial match queries
ACM Transactions on Database Systems (TODS)
Disk allocation for Cartesian product files on multiple-disk systems
ACM Transactions on Database Systems (TODS)
PDIS '93 Proceedings of the second international conference on Parallel and distributed information systems
Latin Squares for Parallel Array Access
IEEE Transactions on Parallel and Distributed Systems
Cyclic Allocation of Two-Dimensional Data
ICDE '98 Proceedings of the Fourteenth International Conference on Data Engineering
A Multiuser Performance Analysis of Alternative Declustering Strategies
Proceedings of the Sixth International Conference on Data Engineering
Disk Allocation Methods for Parallelizing Grid Files
Proceedings of the Tenth International Conference on Data Engineering
Performance Evaluation of Grid Based Multi-Attibute Record Declustering Methods
Proceedings of the Tenth International Conference on Data Engineering
Optimal Allocation of Two-Dimensional Data
ICDT '97 Proceedings of the 6th International Conference on Database Theory
Study of Scalable Declustering Algorithms for Parallel Grid Files
IPPS '96 Proceedings of the 10th International Parallel Processing Symposium
A General Multidimensional Data Allocation Method for Multicomputer Database Systems
DEXA '97 Proceedings of the 8th International Conference on Database and Expert Systems Applications
ICDE '95 Proceedings of the Eleventh International Conference on Data Engineering
Declustering Using Golden Ratio Sequences
ICDE '00 Proceedings of the 16th International Conference on Data Engineering
Distributed computation of the knn graph for large high-dimensional point sets
Journal of Parallel and Distributed Computing
Algorithms and data structures for external memory
Foundations and Trends® in Theoretical Computer Science
Information Sciences: an International Journal
Toward boosting distributed association rule mining by data de-clustering
Information Sciences: an International Journal
Hi-index | 0.00 |
Range queries are an important class of queries for several applications can be achieved by tiling the multi-dimensional data and distributing it among multiple disks or nodes. It has been established that schemes that achieve optimal parallel block access exist only for a few special cases. Though several schemes for the allocation of tiles to disks have been developed, no scheme with non-trivial worst-case bound is known. We establish that any range query on a 2q × 2q-block grid of blocks can be performed using k = 2t disks (t ≤ q), in at most ⌈m/k⌉ + O(logk) parallel block accesses. We achieve this result by judiciously distributing the blocks among the k nodes or disks. Experimental data show that the algorithm achieves very close to ⌈m/k⌉ performance (on average less than 0.5 away from ⌈m/k⌉, with a worst-case of 3). Although several declustering schemes for range queries have been developed, prior to our work no additive non-trivial performance bounds were known. Our scheme guarantees performance within a (small) additive deviation from ⌈m/k⌉. Subsequent to this work, Bhatia et al. [Hierarchical declustering schemes for range queries, in: Proceedings of International Conference on Extending Database Technology, Konstanz, Germany, March 2000, p. 525] have proved that such a performance bound is essentially optimal for this kind of scheme, and have also extended our results to the case where the number of disks is a product of the form k1 * k2 * ... * kt where the kis need not all be 2.