Systems, methods and software for automating database tasks

• US 20060026212A1
• Filed: 07/29/2004
• Published: 02/02/2006
• Est. Priority Date: 07/29/2004
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Tsukerman, Alex, Ramacher, Mark, Voss, Eric P., Draaijer, Jacco, Takahara, Mineharu, Chen, Jane Xin
• Granted Patent: US 7,440,973 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/21 Design, administration or maintenance of databases
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

Various embodiments of the invention provide solutions to facilitate the management and/or automation of various database tasks, in particular system tasks. Merely by way of example, some embodiments provide methods systems and/or software products that can facilitate automation of such tasks. In accordance with certain embodiments, a maintenance window may be defined. The maintenance window can provide an opportunity for system tasks (and/or other tasks, especially those that consume significant database resources) to execute without significantly impacting the usability of the database. In particular embodiments, a maintenance window may be defined to coincide with a period of relatively low user activity. In some cases, a maintenance window may be implemented in conjunction with a job scheduler and/or a resource manager. Merely by way of example, a resource manager might allocate relatively more system and/or database resources to system tasks during the maintenance window, and/or a job scheduler may schedule system tasks to execute during the maintenance window.

Claims

1. In a database environment, a software framework for managing system resources, the software framework comprising:

a maintenance window during which system tasks may be executed without significantly impacting user tasks;

a resource manager configured to establish at least one resource plan, wherein the resource plan defines how system resources should be allocated among a plurality of competing tasks, the plurality of competing tasks comprising at least one user task and at least one system task; and

a job scheduler configured to govern the execution of the at least one system task;

wherein;

the resource manager is configured to allocate relatively more system resources for the at least one system task during the system defined maintenance window; and

the job scheduler is configured to inhibit the at least one system task from executing automatically except during the system-defined maintenance window.

2. In a database environment, a computer-implemented method for managing system resources, the method comprising:

defining a maintenance window during which system tasks may be executed without significantly impacting user tasks;

scheduling at least one system task to execute during the maintenance window, such that the execution of the at least one system task imposes a relatively minimal impact on at least one user task;

allocating relatively greater system resources to the at least one system task during the defined maintenance window; and

inhibiting the at least one system task from executing except during the defined maintenance window.

3. A method for managing system resources as recited in claim 2, the method further comprising:

estimating whether the a duration of the defined maintenance window will be sufficient to allow complete execution of the at least one system task.

4. A method for managing system resources as recited in claim 3, the method further comprising:

issuing a warning if it is estimated that the duration of the defined maintenance window will not be sufficient will not be sufficient to allow complete execution of the at least one system task.

5. A method for managing system resources as recited in claim 3, the method further comprising:

disabling the system task if it is estimated that the duration of the defined maintenance window will not be sufficient will not be sufficient to allow complete execution of the at least one system task.

6. A method for managing system resources as recited in claim 3, the method further comprising:

extending the defined maintenance window if it is estimated that the duration of the defined maintenance window will not be sufficient will not be sufficient to allow complete execution of the at least one system task.

7. A method for managing system resources as recited in claim 2, the method further comprising:

evaluating the at least one system task to determine whether the at least one system task likely will consume sufficient resources to impact user tasks.

8. A method for managing system resources as recited in claim 7, the method further comprising:

determining, based on an evaluation of the at least one system task, that the system task should be inhibited from executing except during the defined maintenance window.

9. A method for managing system resources as recited in claim 2, the method further comprising:

establishing a resource plan that defines how system resources should be allocated among a plurality of competing tasks, the plurality of competing tasks comprising the at least one user task and the at least one system task.

10. A method for managing system resources as recited in claim 9, the method further comprising:

modifying the resource plan to allocate relatively more system resources to the at least one system task during the defined maintenance window.

11. A method for managing system resources as recited in claim 9, wherein the resource plan is a first resource plan, and wherein the first resource plan defines how system resources should be allocated outside the defined maintenance window, the method further comprising:

establishing a second resource plan that defines how system resources should be allocated among the plurality of competing tasks during the defined maintenance window, wherein the second resource plan allocates relatively more system resources to the at least one system task during the defined maintenance window than the first resource plan.

12. A method for managing system resources as recited in claim 2, the method further comprising:

inhibiting the execution of the at least one user task during the defined maintenance window.

13. A method for managing system resources as recited in claim 2, the method further comprising:

allowing a user to schedule a user task to run as a system task during the defined maintenance window.

14. A method for managing system resources as recited in claim 2, wherein defining a maintenance window comprises automatically defining a maintenance window.

15. A method for managing system resources as recited in claim 2, wherein defining a maintenance window comprises:

analyzing a historical pattern of user activity for the database, wherein the historical pattern of user activity comprises at least one periodic segment of relatively high user activity and at least one periodic segment of relatively low user activity; and

defining a maintenance window that coincides with the at least one periodic segment of relatively low user activity.

16. A method for managing system resources as recited in claim 2, wherein defining a maintenance window comprises allowing a user to define a maintenance window.

17. A method for managing system resources as recited in claim 2, wherein defining a maintenance window comprises defining a plurality of maintenance windows.

18. A method for managing system resources as recited in claim 17, wherein the plurality of maintenance windows comprises a first maintenance window and a second maintenance window, the method further comprising:

defining a maintenance window group, the maintenance window group comprising the first maintenance window and the second maintenance window.

19. A method for managing system resources as recited in claim 17, wherein:

the plurality of maintenance windows comprises a first maintenance window and a second maintenance window; and

scheduling at least one system task to execute during the maintenance window comprises;

scheduling a first system task to execute during the first maintenance window; and

scheduling a second system task to execute during the second maintenance window.

20. A method for managing system resources as recited in claim 19, wherein:

the first maintenance window is defined to recur periodically over a first interval, such that the first system task executes periodically over the first interval; and

the second maintenance window is defined to recur periodically over a second interval, such that the second system task executes periodically over the second interval.

21. A method for managing system resources as recited in claim 20, wherein the first interval is not the same as the second interval.

22. A method for managing system resources as recited in claim 20, wherein the first interval is offset from the second interval, such that the first maintenance window and the second maintenance window do not occur at the same time.

23. A method for managing system resources as recited in claim 2, wherein defining a maintenance window comprises defining a maintenance window to recur periodically.

24. A method for managing system resources as recited in claim 2, the method further comprising:

if the at least one system task is has not finished executing before the maintenance window has expired, allowing the at least one system task to finish executing outside the maintenance window.

25. A method for managing system resources as recited in claim 2, the method further comprising:

if the at least one system task is has not finished executing before the maintenance window has expired, stopping the execution of the at least one system task at the expiration of the maintenance window.

26. A method for managing system resources as recited in claim 25, the method further comprising:

rolling back a result of the at least one system task to a state that existed before the execution of the at least one system task.

27. A method for managing system resources as recited in claim 25, the method further comprising:

if the execution of the at least one system task is stopped before the at least one system task has fully executed, saving a state of the at least one system task.

28. A method for managing system resources as recited in claim 27, the method further comprising:

resuming during a subsequent maintenance window the at least one system task, based on the saved state of the at least one system task.

29. A method for managing system resources as recited in claim 25, the method further comprising:

if the execution of the at least one system task is stopped before the at least one system task has fully executed, notifying an administrator that the at least one system task failed to fully execute.

30. A method for managing system resources as recited in claim 25, the method further comprising:

if the execution of the at least one system task is stopped before the at least one system task has fully executed, allowing an administrator to increase the duration of a subsequent maintenance window to allow the at least one system task to fully execute during the subsequent maintenance window.

31. A method for managing system resources as recited in claim 25, the method further comprising:

if the execution of the at least one system task is stopped before the at least one system task has fully executed, automatically increasing the duration of a subsequent maintenance window to allow the at least one system task to fully execute during the subsequent maintenance window.

32. A method for managing system resources as recited in claim 25, wherein the system task comprises an operation selected from the group consisting of a purge operation, a statistic gathering operation, a database analysis operation, a feature usage analysis operation, a backup operation and a database replication operation.

33. A computer system comprising:

a database;

a processor in communication with the database; and

a computer readable medium in communication with the processor, the computer readable medium comprising instructions executable by the processor to;

define a maintenance window for the database during which system tasks may be executed without significantly impacting user tasks;

schedule at least one system task to execute during the maintenance window, such that the execution of the at least one system task imposes a relatively insignificant impact on at least one user task;

allocate relatively greater system resources to the at least one system task during the defined maintenance window; and

inhibit the at least one system task from executing except during the defined maintenance window.

34. A computer program stored on a computer-readable storage medium for managing system resources for a database, the computer program comprising instructions executable by a computer to:

define a maintenance window for the database during which system tasks may be executed without significantly impacting user tasks;

schedule at least one system task to execute during the maintenance window, such that the execution of the at least one system task imposes a relatively insignificant impact on at least one user task;

allocate relatively greater system resources to the at least one system task during the defined maintenance window; and

inhibit the at least one system task from executing except during the defined maintenance window.

35. A computer program as recited in claim 34, wherein the computer program is incorporated in a relational database management system.

36. A computer program as recited in claim 34, wherein the computer program is configured to interoperate with a relational database management system.