EXPLOITING CONDITIONS TO OPTIMIZE EXPENSIVE DATABASE QUERIES

• US 20090259624A1
• Filed: 04/11/2008
• Published: 10/15/2009
• Est. Priority Date: 04/11/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: DeMaris, Lincoln, Fink, Nathan James, Jackson, Anthony Lawrence, Friesenhahn, Dustin
• Granted Patent: US 8,645,406 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24534 Query rewriting; Transformation

Abstract

A fallback query mechanism that serves as a way to recover from query conditions (or criteria) that break due to excessively large result sets relative to a preset threshold. The fallback query mechanism detects when a query has failed to execute because the query requested more items from the database (or list) than allowed by the threshold, rewrites the query to examine no more than the threshold number of most recently created items that match one of the indexed filter criteria specified in a view, and then applies one or more of the other filter criteria. The mechanism then reruns the query and displays the results to the user.

Claims

1. A computer-implemented query system, comprising:

a detection component for detecting when an original query fails due to conditions imposed on a data source against which the original query is run, the conditions related to performance of the data source; and

a query component for automatically creating a new query based on the original query that when executed, the new query returns a new result set that meets the conditions.

2. The system of claim 1, wherein the detection component and the query component are part of an enterprise web-based collaboration and data management server that runs the original query against the data source, which is an enterprise database.

3. The system of claim 1, wherein the query component creates the new query such that when executed, the new query returns the new result set based on conditions imposed on the original query.

4. The system of claim 3, wherein the conditions include one or more of system conditions that comprise a threshold relative to a number of items to return, or user-specified conditions that comprise filters, groups, or sorts.

5. The system of claim 1, wherein the performance of the data source is related to query execution processing and results processing that affects overall quality of service of the data source.

6. The system of claim 1, wherein the new query is created based on user-selected criteria that is automatically run.

7. The system of claim 1, wherein the new result set is a subset of the original result set and includes most recently created documents from the original result set.

8. The system of claim 1, further comprising an optimization component for optimizing results of the new result set by selecting one or more terms from the original query to employ in the new query, the selection based on optimizing quality of the new result set without triggering a condition.

9. The system of claim 1, further comprising an optimization component for optimizing results of the new result set by inserting one or more terms into the new query, the one or more terms related to a characteristic about an entity generating the original query and which optimize quality of the new result set without triggering a condition.

10. A computer-implemented query system, comprising:

a detection component of a web-based collaboration server for detecting a failure of an original query due an excessive number of items in an original result set that triggers a threshold condition; and

a query component of the web-based collaboration server for automatically rewriting the original query into a new query in response to the detected failure, the new query returns a new result set that does not trigger the threshold condition.

11. The system of claim 10, wherein the new query searches the most recent items of the original result set to reduce the number of items of original result set to the new result set.

12. The system of claim 10, wherein the new query searches the items of the original result set to return the new result set, the new query employs one or more terms associated with a filter, a group or a sort.

13. The system of claim 10, wherein the query component selectively combines indices of terms in the original query to determine an optimal new query that returns the new result set.

14. A computer-implemented method of processing a query, comprising:

detecting a failure of an original query due to triggering of a condition imposed on a data source against which the original query is run, the condition related to performance of the data source;

rewriting the original query into a new query in response to the failure, the new query based on one or more terms of the original query; and

executing the new query against the original result set to return a new result set that meets the condition.

15. The method of claim 14, further comprising rewriting the original query into the new query such that the new result set includes a number of items that does not exceed a maximum-number-of-items threshold condition.

16. The method of claim 14, wherein the new query returns results in the new result set that match terms defined in the original query.

17. The method of claim 14, further comprising building the new query based on compound indices of the terms of the original query.

18. The method of claim 14, further comprising applying filtering and ordering associated with the original query to the new query.

19. The method of claim 14, wherein the new query is an indexed top fallback query, a simple fallback query, or an ID limit fallback query.

20. The method of claim 14, further comprising rewriting the original query into the new query based on a type of the condition imposed on the data source.