Configurable reuse delay criterion for storage volumes

• US 5,644,698 A
• Filed: 05/30/1996
• Issued: 07/01/1997
• Est. Priority Date: 05/30/1996
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Cannon, David Maxwell
• Primary Examiner(s): Beausoliel, Jr., Robert W.
• Assistant Examiner(s): Snyder, Glenn
• Application Number:
  US08/655,526
• Time in Patent Office: 397 Days
• Field of Search: 395/182.04, 395/182.13, 395/182.18
• US Class Current: 714/6.31
• CPC Class Codes: G06F 11/1451 by selection of backup contents
• CPC Class Codes: G06F 11/1461 Backup scheduling policy

Abstract

A data processing system using a storage management server provides a method and apparatus for preserving consistency between a database back-up and a set of storage volumes. The system includes a plurality of client systems coupled to the storage management server. Primary and back-up copies of client files are stored within a set of attached storage volumes. The server contains a server database to maintain directory and reference location information linking the primary and back-up copies of the client files. The server periodically performs an incremental back-up of the server database to a database back-up. The server utilizes a reuse delay criterion to set selected storage volumes as pending volumes. Pending volumes are not eligible for reuse by the server, thereby preventing the server from overwriting these pending volumes and invalidating reference location information for the pending volumes stored in the database back-up. When the reuse delay criterion elapses, the pending volumes are classified as empty, making them eligible for reuse by the server.

Claims

1. In a data processing system having a plurality of client systems generating client data files, and a storage management server coupled to said plurality of client systems, said server having a plurality of storage volumes for storing a primary copy and a back-up copy of said client data files, a server database for maintaining directory and reference location information linking said primary copy and said back-up copy of said client data files, and a database back-up for storing an incremental back-up of said server database, a method in said server for preserving consistency between said database back-up and said plurality of storage volumes comprising steps of:

(a) maintaining storage volume information in said server database, an entry in said storage volume information corresponding to each of said plurality of storage volumes;

(b) specifying a reuse delay criterion for a selected storage volume;

(c) marking said selected storage volume as a pending storage volume when a last remaining file is removed from said selected storage volume to prevent said server from invalidating references within said database back-up by overwriting said selected storage volume;

(d) periodically checking whether said reuse delay criterion has elapsed for said pending storage volume; and

(e) classifying said pending storage volume as eligible for reuse by said server when said reuse delay criterion has elapsed.

2. The method of claim 1 further comprising:

(f) marking said selected storage volume as an empty storage volume when said reuse delay criterion is not specified for said selected storage volume.

3. The method of claim 2 wherein the step (b) further comprises setting a reuse delay field in said storage volume information entry corresponding to said selected storage volume.

4. The method of claim 3 wherein the step (c) further comprises:

setting a volume status field to a pending status in said storage volume information entry corresponding to said selected storage volume; and

indicating when said selected volume was initially set pending in a starting criterion field in said storage volume information entry corresponding to said selected storage volume.

5. The method of claim 4 wherein the step (d) further comprises:

calculating a current delay criterion for said pending storage volume, said current delay criterion being the difference between a current criterion and said starting criterion; and

comparing said current delay criterion with said reuse delay criterion corresponding to said pending storage volume.

6. The method of claim 5 wherein the step (e) further comprises:

setting said volume status field to an empty status in said storage volume information entry corresponding to said pending storage volume; and

clearing said starting criterion field in said storage volume information entry corresponding to said selected storage volume.

7. The method of claim 6 wherein the step (f) further comprises setting said volume status field to said empty status in said storage volume information entry corresponding to said selected storage volume.

8. The method of claim 7 wherein said reuse delay criterion comprises an amount of elapsed time from when said last remaining file is removed from said selected storage volume until said server can reuse said selected storage volume.

9. The method of claim 8 wherein said reuse delay criterion comprises a number of incremental back-ups to occur before said server can reuse said selected storage volume.

10. The method of claim 9 wherein said reuse delay criterion comprises a command to be issued to said server before said server can reuse said selected storage volume.

11. In a data processing system having a plurality of client systems generating client data files, and a storage management server coupled to said plurality of client systems, said server having a plurality of storage volumes for storing a primary copy and a back-up copy of said client data files, said server comprising:

a server database for maintaining directory and reference location information linking said primary copy and said back-up copy of said client data files, said server also maintaining a storage volume list having an information entry which includes a reuse delay criterion, corresponding to each of said plurality of storage volumes;

a database back-up for storing an incremental back-up of said server database, said database back-up also maintaining directory, reference location, and storage volume information pertaining to said storage volumes; and

a storage manager coupled to said plurality of storage volumes, said server database, and said database back-up, said storage manager storing said primary and said back-up copy of said client data files in said plurality of storage volumes, said storage manager also periodically performing an incremental back-up operation from said server database to said database back-up, said storage manager further preserving consistency between said database back-up and said plurality of storage volumes by marking a selected storage volume as a pending storage volume when a last remaining copy of said client data files is removed from said selected storage volume, thereby preventing said server from invalidating reference location within said database back-up by overwriting said selected storage volume.

12. The server in claim 11 further comprising:

a volume status field within said information entry in said storage volume list; and

a starting criterion within said information entry in said storage volume list.

13. The server in claim 12 wherein said storage manager sets said volume status according to said reuse delay criterion corresponding to said selected storage volume.

14. The server in claim 13 wherein said storage manager sets a pending status in said volume status field corresponding to said selected storage volume having said reuse delay criterion, said pending status indicating said selected storage volume is a pending storage volume, not eligible for reuse by said server.

15. The server in claim 14 further comprising:

a first list in said server database identifying said pending storage volumes,

wherein said storage manager queries said storage volume list to build said first list.

16. The server in claim 15 wherein, for each said pending storage volume, said storage manager periodically determines whether said reuse delay criterion have elapsed.

17. The server in claim 15 wherein said storage manager sets an empty status in said volume status field corresponding to said selected storage volume not having said reuse delay criterion, said empty status indicating said selected storage volume is eligible for reuse by said server.

18. The server in claim 17 wherein said storage manager sets said empty status in said volume status field corresponding to said pending storage volume once said reuse delay criterion has elapsed, indicating said pending storage volume is now eligible for reuse by said server.

19. The server in claim 18 wherein said reuse delay criterion comprises an amount of elapsed time from when said last remaining copy is moved or deleted from said selected storage volume until said server can reuse said selected storage volume.

20. The server in claim 18 wherein said reuse delay criterion comprises a number of incremental back-ups to occur before said server can reuse said selected storage volume.

21. The server in claim 20 wherein said reuse delay criterion comprises a command to be issued to said server before said server can reuse said selected storage volume.

22. In a data processing system having a plurality of client systems generating client data files, and a storage management server coupled to said plurality of client systems, said server having a plurality of storage volumes for storing a primary copy and a back-up copy of said client data files, a server database for maintaining directory and reference location information linking said primary copy and said back-up copy of said client data files, and a database back-up for storing an incremental back-up of said server database, a program product having executable computer instructions for preserving consistency between said database back-up and said plurality of storage volumes comprising:

a computer readable storage medium for storing said executable computer instructions, said executable computer instructions comprising;

maintaining storage volume information in said server database, an entry in said storage volume information corresponding to each of said plurality of storage volumes;

specifying a reuse delay criterion for a selected storage volume;

marking said selected storage volume as a pending storage volume when a last remaining file is removed from said selected storage volume to prevent said server from invalidating references within said database back-up by overwriting said selected storage volume;

periodically checking whether said reuse delay criterion has elapsed for said pending storage volume;

classifying said pending storage volume as eligible for reuse by said server when said reuse delay criterion has elapsed; and

marking said selected storage volume as an empty storage volume when said reuse delay criterion is not specified for said selected storage volume.