Persistent querying in a federated database system

• US 8,386,515 B2
• Filed: 11/23/2010
• Issued: 02/26/2013
• Est. Priority Date: 11/27/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bent, Graham, Gibson, Christopher R., Vyvyan, David R., Dantressangle, Patrick
• Primary Examiner(s): ROBINSON, GRETA LEE
• Application Number:
  US12/952,803
• Publication Number:
  US 20110131242A1
• Time in Patent Office: 826 Days
• Field of Search: 707/627, 707/628, 707/629, 707/630, 707/631, 707/632, 707/633, 707/705, 707/764, 707/770, 707/766, 707/767, 707/773, 707/774, 707/775, 707/999.002, 707/999.01
• US Class Current: 707/775
• CPC Class Codes: G06F 16/2471 Distributed queries

Abstract

Methods and systems are provided, for persistent querying in a federated database system. A federated database system has a plurality of connectable database management system nodes, each database node having at least one database. Persistent querying includes: receiving a query at a first database node and propagating the query from the first database node to other database nodes in the federated database system. A responding database node satisfying the query: obtains a query results set; defines a topic identifier for the query; and publishes subsequent messages relating to the topic identifier when data updates are available. The first database node receives query results sets from one or more responding nodes and the first database node subscribes to topic identifiers for publications of subsequent messages from the responding database nodes. Subsequent messages are published relating to the query along the connection through which the query is returned.

Claims

1. A method for persistent querying in a federated database system having a plurality of connectable database management system nodes, each database node having at least one database, the method comprising:

receiving a query at a first database node;

propagating the query from the first database node to other database nodes in the federated database system;

receiving, by the first database node, query results sets from at least one responding node, wherein at least one responding node satisfying the query;

obtains a query results set;

defines a topic identifier for the query; and

publishes subsequent messages relating to the topic identifier when data updates are available;

subscribing, by the first database node, to topic identifiers for publications of subsequent messages from the responding database nodes; and

optimizing a route from the responding node to the first database node by periodically re-issuing the query from the first database node to other database nodes in the federated database system and wherein the responding node further defines a new topic identifier associated with the re-issued query.

2. The method as claimed in claim 1, further comprising:

publishing the subsequent messages relating to the query along the connection through which the query is returned.

3. The method as claimed in claim 1, wherein:

receiving query results sets further comprises receiving, by the first database node, the query results sets from an intermediate responding node that aggregates query results sets and topic identifiers from lower responding database nodes in the propagation.

4. The method as claimed in claim 1, further comprising:

decomposing the query into sub-queries before propagation to other database nodes;

wherein a corresponding topic identifier relates to an associated sub-query.

5. The method as claimed in claim 1, further comprising:

defining the query as a persistent query to inform the database nodes that publication of subsequent messages is required.

6. The method as claimed in claim 1, further comprising:

instructing a message broker to subscribe to messages on its topic identifier and to publish subsequent message to other database nodes.

7. The method as claimed in claim 1, further comprising:

enabling a network of message brokers on aggregated topic identifiers in a route of returned query results from database nodes.

8. The method as claimed in claim 7, further comprising:

recording return routes of topic identifiers; and

enabling message brokers on the optimum return routes.

9. The method as claimed in claim 1, further comprising:

receiving by a select one responding database node, the same query from another node; and

identifying existing topic identifiers that satisfy the query.

10. The method as claimed in claim 1, further comprising:

storing topic identifiers locally on a database node for use with future queries.

11. The method as claimed in claim 10, further comprising:

storing meta-data with topic identifiers to aid discoverability of a topic identifier for a query.

23. The method of claim 1, wherein the topic identifier and the new topic identifier are identical.

12. A computer-readable storage device with an executable program for persistent querying in a federated database system having a plurality of connectable database management system nodes, each database node having at least one database, wherein the program instructs a processor to perform:

receiving a query at a first database node;

propagating the query from the first database node to other database nodes in the federated database system;

receiving, by the first database node, query results sets from at least one responding node, wherein at least one responding node satisfying the query;

obtains a query results set;

defines a topic identifier for the query; and

publishes subsequent messages relating to the topic identifier when data updates are available;

subscribing, by the first database node, to topic identifiers for publications of subsequent messages from the responding database nodes;

optimizing a route from the responding node to the first database node by periodically re-issuing the query from the first database node to other database nodes in the federated database system and wherein the responding node further defines a new topic identifier associated with the re-issued query.

13. A system for persistent querying in a federated database system with a plurality of connectable database management system nodes, each database node having at least one database, comprising:

a processor coupled to a memory such that the processor is programmed to implement;

a query mechanism for receiving a query at a first database node;

a propagating mechanism for propagating the query from the first database node to other database nodes in the federated database system;

a responding database node including;

a query results set obtaining component; and

a topic identifier defining component for the query; and

a publisher application for publishing subsequent messages relating to the topic identifier when data updates are available;

wherein;

the first database node includes a results receiving component for query results sets from one or more responding nodes;

the first database node includes a subscriber application for subscribing to topic identifiers for publications of subsequent messages from the responding database nodes; and

the first database node optimizes a route from the responding node to the first database node by periodically re-issuing the query from the first database node to other database nodes in the federated database system and wherein the responding node further defines a new topic identifier associated with the re-issued query.

14. The system as claimed in claim 13, wherein:

an intermediate responding node includes an aggregating component to aggregate query results sets and topic identifiers from lower responding database nodes in the propagation.

15. The system as claimed in claim 13, wherein:

the first database node includes a query decomposition component for decomposing a query into sub-queries before propagation to other database nodes and the topic identifier relates to the sub-query.

16. The system as claimed in claim 13, including a definition that the query is a persistent query to inform the database nodes that publication of subsequent messages is required.

17. The system as claimed in claim 13, wherein:

a database node instructs a message broker to subscribe to messages on its topic identifier and to publish subsequent message to other database nodes.

18. The system as claimed in claim 13, wherein:

a network of message brokers is enabled on the aggregated topic identifiers in a route of returned query results from database nodes.

19. The system as claimed in claim 13, including a route manager component wherein return routes of topic identifiers are recorded and message brokers enabled on the optimum return routes.

20. The system as claimed in claim 13, wherein:

a responding database node includes an existing topic identifier component for identifying existing topic identifiers that satisfy a query received from another node.

21. The system as claimed in claim 13, wherein:

a database node includes a table for storing topic identifiers locally for use with future queries.

22. The system as claimed in claim 21, wherein:

the table includes meta-data stored with topic identifiers to aid discoverability of a topic identifier for a query.