Apparatus and method for analyzing query optimizer performance

1. A method of analyzing the performance of a query optimizer, comprising:

interrogating in response to an event trigger software components of a query optimizer of a customer computer to collect serialized data, said event trigger being selected from a query optimizer crash, the submission of a specified query, and a request for a reproduction object;

populating at the customer computer a reproduction object characterizing operational parameters of the customer computer at the time of the event trigger, the operational parameters comprising said serialized data, query and database information, and system information characterizing physical machines that process a query execution plan object, the operational parameters further including an input query, a query plan object, database schema information, database statistics and physical machine data;

transporting the reproduction object from the customer computer to a test computer; and

analyzing the reproduction object at the test computer to characterize the performance of the query optimizer.

2. The method of claim 1 wherein analyzing includes analyzing a query optimizer crash.

3. The method of claim 1 wherein analyzing includes analyzing a suboptimal query plan.

4. The method of claim 1 wherein analyzing includes analyzing an incorrect query plan.

5. The method of claim 1 wherein analyzing includes analyzing a query optimization workflow.

6. The method of claim 1 wherein analyzing includes performing a what-if scenario.

7. A customer computer, comprising:

a processor; and

a memory connected to the processor, the memory storing a collection module including instructions to cause the processor to:

interrogate in response to an event trigger software components of a query optimizer of the customer computer to collect serialized data,

populate a reproduction object characterizing operational parameters of the customer computer at the time of the event trigger, the operational parameters comprising said serialized data, query and database information, and system information characterizing physical machines that process a query execution plan object, and

transport the reproduction object from the customer computer to a test computer for query optimizer performance analysis, wherein the event trigger is selected from a query optimizer crash, the submission of a specified query and a request for a reproduction object, and the operational parameters include an input query, a query plan object, database schema information, database statistics and physical machine data.

8. A test computer, comprising:

a processor; and

a memory connected to the processor, the memory storing an analysis module including instructions to cause the processor to:

analyze a reproduction object characterizing operational parameters of a customer computer at the time of an event trigger to characterize the performance of a query optimizer of the customer computer, the operational parameters comprising serialized data from software components of said query optimizer, query and database information, and system information characterizing physical machines that process a query execution plan object, and wherein the operational parameters further include an input query, a query plan object, database schema information, database statistics and physical machine data, and wherein said event trigger includes a query optimizer crash.

9. The test computer of claim 8 wherein the executable instructions to analyze include executable instructions to analyze a suboptimal query plan.

10. The test computer of claim 8 wherein the executable instructions to analyze include executable instructions to analyze an incorrect query plan.

11. The test computer of claim 8 wherein the executable instructions to analyze include executable instructions to analyze a query optimization workflow.

12. The test computer of claim 8 wherein the executable instructions to analyze include executable instructions to analyze a what-if scenario.