Efficiently Processing Query-Intensive Databases over a Non-Dedicated Local Network
IPDPS '05 Proceedings of the 19th IEEE International Parallel and Distributed Processing Symposium (IPDPS'05) - Papers - Volume 01
Hierarchical aggregation in networked data management
Euro-Par'05 Proceedings of the 11th international Euro-Par conference on Parallel Processing
Hi-index | 0.00 |
This paper discusses efficient hash-partitioning usingworkload access patterns to place and process relationsin a cluster or distributed query-intensive databaseenvironment.In such an environment, there is usuallymore than one partitioning alternative for each relation.We discuss a method and algorithm to determine the haspartitioning attributes and placement.Among thealternatives, our algorithm chooses a placement thatreduces repartitioning overheads using expected orhistorical query workloads.The paper includes asimulation study showing how our strategy outperformsad-hoc placement and previously proposed distributeddatabase strategies.