Techniques for data assignment from an external distributed file system to a database management system
AssigneeTeradata US, Inc.
InventorsYan Qi, Yu Xu, Olli Pekka Kostamaa, Jian Wen
Life of this patent
Estimating the term…
Abstract
Techniques for data assignment from an external distributed file system (DFS) to a database management system (DBMS) are provided. Data blocks from the DFS are represented as first nodes and access module processors of the DBMS are represented as second nodes. A graph is produced with the first and second nodes. Assignments are made for the first nodes to the second nodes based on evaluation of the graph to integrate the DFS with the DBMS.
Claims20
A method implemented and programmed within a non-transitory computer-readable storage medium and processed by one or more processors, the processors configured to execute the method, comprising:
- receiving an initial assignment of first nodes to second nodes in a graph, the first nodes representing data blocks in an external distributed file system and the second nodes representing access module processors of a database management system (DBMS);
- constructing a residual graph with a negative cycle having an initial assignment;
- iterating the residual graph and with each iteration adjusting the initial assignment by any present negative cycle for the residual graph until there are no negative cycles present in the residual graph;
- and
- returning a final assignment for each of the data blocks to one of the access module processors as an assignment flow.
A method implemented and programmed within a non-transitory computer-readable storage medium and processed by one or more processors, the processors configured to execute the method, comprising:
- obtaining data blocks for an external distributed file system;
- acquiring access module processors for a database management system (DBMS);
- organizing a first set of nodes to represent the data blocks and a second set of nodes as the access module processors within a bipartite graph;
- using the first set of nodes and the second set of nodes to produce a minimum cost bipartite graph with each of the first set of nodes assigned to a specific one of the second nodes in the second set of nodes;
- and
- providing the minimum cost bipartite graph as a final assignment for the first set of nodes mapped to the second set of nodes.
A method implemented and programmed within a non-transitory computer-readable storage medium and processed by one or more processors, the processors configured to execute the method, comprising:
- generating a graph having a source node, first nodes, second nodes, and a target node;
- representing each first node as a block of data from an external file system and each second node as an access module processor on a database management system;
- processing an approximate-greedy algorithm on the source node, the first nodes, the second nodes, and the target node to produce a modified graph having assignments between the first nodes and the second nodes;
- and
- returning a pointer to the graph.
Dependencies are inferred from the claim text.