Storage and Access Methods for Advanced Traveler Information Systems
Shashi Shekhar, Duen-Ren Liu
Report no. Mn/DOT 1996-26
Topics: Traffic Modeling and Data
This report examines the use of the Connectivity-Clustered Access Method (CCAM) to improve network operations. The expected I/O cost for many network operations can be reduced by maximizing the Weighted Connectivity Residue Ratio (WCRR), i.e., the chance that a pair of connected nodes that are more likely to be accessed together are allocated to a common page of the file. An access method for general networks that uses connectivity clustering, CCAM supports the operations of insert, delete, create, and find, as well as the new operations, get-A-successor and get-successors, which retrieve one or all successors of a node to facilitate aggregate computations on networks. The nodes of the network are assigned to disk pages via a graph partitioning approach to maximize the WCRR. CCAM includes methods for static clustering, as well as dynamic incremental reclustering, to maintain high WCRR in the face of updates, without incurring high overheads. The report also describes possible modifications to improve the WCRR that can be achieved by existing spatial access methods. Experiments with network computations on the Minneapolis road map show that CCAM outperforms existing access methods, even though the proposed modifications also substantially improve the performance of existing spatial access methods.