Transaction-aware caching for document metadata

• US 20030200197A1
• Filed: 05/30/2003
• Published: 10/23/2003
• Est. Priority Date: 05/12/2000
• Status: Active Grant

• Current Assignee: Oracle International Corporation (Oracle Corporation)
• Original Assignee: Oracle International Corporation (Oracle Corporation)
• Inventors: Pitfield, David B., Long, David J.
• Granted Patent: US 7,987,217 B2
• Time in Patent Office: Days
• US Class Current: 707/1
• CPC Class Codes: H04L 63/10 for controlling access to devices or network resources
• CPC Class Codes: H04L 67/06 specially adapted for file transfer, e.g. file transfer protocol [FTP]
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

Techniques are provided for performing transaction-aware caching of metadata in an electronic file system. A mechanism is described for providing transaction-aware caching that uses a cache hierarchy, where the cache hierarchy includes uncommitted caches associated with sessions in an application and a committed cache that is shared among the sessions in that application. Techniques are described for caching document metadata, access control metadata and folder path metadata. Also described is a technique for using negative cache entries to avoid unnecessary communications with a server when applications repeatedly request non-existent data.

Claims

1. A method for finding information about a file that is managed by a database system, comprising the computer-implemented steps of:

receiving a request associated with a session for an operation on the file that is managed by a database system;

looking for an entry associated with said file in a session cache that is associated with said session;

if said entry is not in said session caches, then

looking for said entry in one or more shared caches whose entries are accessible to multiple sessions; and

if said entry is not in said one or more shared caches, then looking in said database for said information about said file.

2. The method of claim 1 wherein said session cache and said one or more shared caches are outside of said database system.

3. The method of claim 1 wherein said session cache contains only entries for files whose modifications are made within said session.

4. The method of claim 1 further comprising, the step of maintaining, within said one or more shared caches, entries only for information that has been committed to said database.

5. The method of claim 1 further comprising the step of when said information about said file is modified in response to an operation performed during said session, placing an entry for said file in said session cache without placing an entry for said modified information in said one or more shared caches.

6. The method of claim 5 wherein said operation modifies said file.

7. The method of claim 5 wherein said operation modifies metadata associated with said file.

8. The method of claim 1 wherein:

the request is generated in response to said file being referenced;

the entry is not in said one or more shared caches; and

the method further includes the step of placing an entry for said information in said one or more shared caches in response to said file being referenced.

9. The method of claim 1 wherein:

the request is generated in response to metadata, which is associated with said file, being referenced;

the entry is not in said one or more shared caches; and

the method further includes the step of placing an entry for said information in said one or more shared caches in response to said metadata being referenced.

10. The method of claim 1 wherein the step of looking for an entry includes looking for an entry that contains attribute metadata associated with said file.

11. The method of claim 10 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes an ID for said file.

12. The method of claim 10 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a name for said file.

13. The method of claim 10 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes an owner for said file.

14. The method of claim 10 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a date said file was created.

15. The method of claim 10 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a date said file was last modified.

16. The method of claim 1 wherein:

said session is associated with a transaction;

the method further comprises the step of;

when said transaction successfully terminates, invalidating entries, in said one or more shared caches, that are for the same information as entries in said session cache.

17. The method of claim 16 further comprising the step of:

deallocating said session cache when said transaction successfully terminates.

18. The method of claim 1 further comprising the steps of:

maintaining a first session cache of said session cache for a first transaction; and

maintaining a second session cache of said session cache for a second transaction;

wherein said second transaction is nested inside of said first transaction.

19. The method of claim 18 further comprising the step of:

deallocating said second session cache when said second transaction successfully terminates without deallocating said first session cache.

20. A method for maintaining an entry in a cache hierarchy comprising the steps of:

storing within a first cache a first version of said entry that includes only changes made by a transaction associated with said first cache;

storing within a second cache a second version of said entry that does not include said changes made by said transaction;

providing data from said first version of said entry to said transaction;

providing data from said second version of said entry to transactions other than said transaction until said transaction successfully terminates; and

invalidating said second version of said entry when said transaction successfully terminates.

21. The method of claim 20 wherein the step of storing within a second cache a second version of said entry includes storing said second version of said entry in a committed cache that only contains entries for data that has been committed to a database.

22. The method for claim 20 wherein said step of invalidating is performed when said transaction commits to a database.

23. The method of claim 20 wherein said transaction is a first transaction that includes a second transaction nested in the first transaction, wherein said first cache is associated with said first transaction and said second cache is associated with said second transaction.

24. The method of claim 23 further comprising the step of:

deallocating said second cache when said second transaction successfully terminates without deallocating the first cache.

25. The method of claim 20 wherein said first cache and said second cache contain data from a database system about files managed by the database system, and said first cache and said second cache are outside of said database system.

26. The method of claim 20 wherein said changes made by said transaction associated with said first cache are not committed to a database until said transaction successfully terminates.

27. The method of claim 20 wherein said second cache only contains entries for files that are committed to a database.

28. A method for maintaining information in a cache hierarchy, the method comprising the steps of:

maintaining a plurality of first caches, wherein each cache of said plurality of caches is associated with a corresponding transaction of a plurality of transactions, and wherein the entries in each first cache contain data that reflects changes made by said corresponding transaction;

maintaining a second cache concurrent with said plurality of first caches, wherein no entries in said second cache contain data that reflects changes made by any of said plurality of transactions; and

in response to successful termination of a given transaction of said plurality of transactions, replacing one or more entries in said second cache with one or more entries from the first cache associated with said given transaction.

29. The method of claim 28 wherein

the step of maintaining a plurality of first caches includes maintaining outside of a database system a plurality of first caches that contain data about files managed by the database system;

and

the step of maintaining a second cache includes maintaining outside of a database system a second cache that contains data about files managed by the database system.

30. The method of claim 28 wherein said given transaction is a first transaction that is nested in a second transaction and said second cache is associated with said second transaction.

31. The method of claim 29 further comprising the step of placing, in the first cache associated with the given transaction, an entry that contains metadata about a file when the file is modified by said given transaction.

32. The method of claim 29 further comprising the step of placing, in the first cache associated with the given transaction, an entry that contains metadata about a file when the metadata is modified by said given transaction.

33. A method for finding metadata comprising the steps of:

maintaining outside of a database system a first cache and a second cache that store said metadata about files stored in said database system;

wherein said first cache stores a first set of metadata that reflects modifications made during a first session and does not reflect modifications made during a second session;

wherein said second cache stores a second set of metadata that reflects said modifications made during said second session and does not reflect said modifications made during said first session;

providing requests received in said first session with data from said first set of metadata; and

providing requests received in said second session with data from said second set of metadata.

34. The method for claim 33 further comprising the steps of:

receiving a request associated with said first session for an operation related to a file stored in the database system;

in response to said request, looking in said first cache for particular metadata associated with said file; and

if said first cache does not have said particular metadata associated with said file, then looking in a third cache for said particular metadata associated with said file,

wherein said third cache only contains file metadata that has been committed by the database system.

35. The method of claim 34 further comprising the step of sending a request for said particular metadata to said database system if said particular metadata is not in said third cache.

36. A computer-readable medium carrying instructions for finding an entry in a cache hierarchy, the instructions comprising instructions for performing the steps of:

receiving a request for an operation associated with a session on a file that is managed by a database system;

looking for said entry associated with said file in session cache that are associated with said session;

if said entry is not in said session cache, then

looking for said entry in one or more shared caches whose entries are accessible to multiple sessions; and

if said entry is not in said one or more shared caches, then looking in said database for said entry.

37. The computer-readable medium of claim 36 wherein said one or more caches and said one or more shared caches are outside of said database system.

38. The computer-readable medium of claim 36 wherein said session cache contain only entries for files whose modifications are made by said session.

39. The computer-readable medium of claim 38 wherein said modifications made by said session are not committed to said database until said session terminates successfully.

40. The computer-readable medium of claim 36 wherein said entries that are accessible to said multiple sessions only cache data that has been committed to a database.

41. The computer-readable medium of claim 36 further comprising instructions for performing the step of placing said entry in said session cache when said file is modified by said session.

42. The computer-readable medium of claim 36 further comprising instructions for performing the step of placing said entry in said session cache when metadata associated with said file is modified by said session.

43. The computer-readable medium of claim 36 wherein:

the request is generated in response to said file being referenced;

the entry is not in said one or more shared caches; and

the computer-readable medium further includes instructions for performing the step of placing an entry for said information in said one or more shared caches in response to said file being referenced.

44. The computer-readable medium of claim 36 wherein:

the request is generated in response to metadata, which is associated with said file, being referenced;

the entry is not in said one or more shared caches; and

the computer-readable medium further includes instructions for performing the step of placing an entry for said information in said one or more shared caches in response to said metadata being referenced.

45. The computer-readable medium of claim 36 wherein the step of looking for an entry includes looking for an entry that contains attribute metadata associated with said file.

46. The computer-readable medium of claim 45 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes an ID for said file.

47. The computer-readable medium of claim 45 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a name for said file.

48. The computer-readable medium of claim 45 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes an owner for said file.

49. The computer-readable medium of claim 45 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a date said file was created.

50. The computer-readable medium of claim 45 wherein the step of looking for an entry that contains attribute metadata includes looking for an entry that includes a date said file was last modified.

51. The computer-readable medium of claim 36 wherein:

said session is associated with a transaction;

the computer-readable medium further comprises instructions for performing the step of;

when said transaction successfully terminates, invalidating entries, in said one or more shared caches, that are for the same information as entries in said session cache.

52. The computer-readable medium of claim 51 further comprising instructions for performing the step of:

deallocating said session cache when said transaction successfully terminates.

53. The computer-readable medium of claim 36 further comprising instructions for performing the steps of:

maintaining a first session cache of said session cache for a first transaction; and

maintaining a second session cache of said session cache for a second transaction;

wherein and said second transaction is nested inside of said first transaction.

54. The computer-readable medium of claim 53 further comprising instructions for performing the step of:

deallocating said second session cache when said second transaction successfully terminates without deallocating said first session cache.

55. A computer-readable medium carrying instructions for maintaining an entry in a cache hierarchy, the instructions comprising instructions for performing the steps of:

storing within a first cache a first version of said entry that includes only changes made by a transaction associated with said first cache;

storing within a second cache a second version of said entry that does not include said changes made by said transaction;

providing data from said first version of said entry to said transaction;

providing data from said second version of said entry to transactions other than said transaction until said transaction successfully terminates; and

invalidating said second version of said entry when said transaction successfully terminates.

56. The computer-readable medium of claim 55 wherein the step of storing within a second cache a second version of said entry includes storing said second version of said entry in a committed cache that only contains entries for data that has been committed to a database.

57. The computer-readable medium for claim 55 wherein said step of invalidating is performed when said transaction commits to a database.

58. The computer-readable medium of claim 55 wherein said transaction is a first transaction that includes a second transaction nested in the first transaction, wherein

said first cache is associated with said first transaction and said second cache is associated with said second transaction.

59. The computer-readable medium of claim 58 further comprising instructions for performing the step of:

deallocating said second cache when said second transaction successfully terminates without deallocating the first cache.

60. The computer-readable medium of claim 55 wherein said first cache and said second cache contain data from a database system about files managed by the database system, and said first cache and said second cache are outside of said database system.

61. The computer-readable medium of claim 55 wherein said changes made by said transaction associated with said first cache are not committed to a database until said transaction successfully terminates.

62. The computer-readable medium of claim 55 wherein said second cache only contains entries for files that are committed to a database.

63. A computer-readable medium for maintaining information in a cache hierarchy, the computer-readable medium comprising instructions for performing the steps of:

maintaining a plurality of first caches, wherein each cache of said plurality of caches is associated with a corresponding transaction of a plurality of transactions, and wherein the entries in each first cache contain data that reflects changes made by said corresponding transaction;

maintaining a second cache concurrent with said plurality of first caches, wherein no entries in said second cache contain data that reflects changes made by any of said plurality of transactions; and

in response to successful termination of a given transaction of said plurality of transactions, replacing one or more entries in said second cache with one or more entries from the first cache associated with said given transaction.

64. The computer-readable medium of claim 63 wherein

the step of maintaining a plurality of first caches includes maintaining outside of a database system a plurality of first caches that contain data about files managed by the database system;

and

the step of maintaining a second cache includes maintaining outside of a database system a second cache that contains data about files managed by the database system.

65. The computer-readable medium of claim 63 wherein said given transaction is a first transaction that is nested in a second transaction and said second cache is associated with said second transaction.

66. The computer-readable medium of claim 64 further comprising instructions for performing the step of placing, in the first cache associated with the given transaction, an entry that contains metadata about a file when the file is modified by said given transaction.

67. The computer-readable medium of claim 64 further comprising instructions for performing the step of placing, in the first cache associated with the given transaction, an entry that contains metadata about a file when the metadata is modified by said given transaction.

68. A computer-readable medium for finding metadata comprising instructions for performing the steps of:

maintaining outside of a database system a first cache and a second cache that store said metadata about files stored in said database system;

wherein said first cache stores a first set of metadata that reflects modifications made during a first session and does not reflect modifications made during a second session;

wherein said second cache stores a second set of metadata that reflects said modifications made during said second session and does not reflect said modifications made during said first session;

providing requests received in said first session with data from said first set of metadata; and

providing requests received in said second session with data from said second set of metadata.

69. The computer-readable medium for claim 68 further comprising instructions for performing the steps of:

receiving a request associated with said first session for an operation related to a file stored in the database system;

in response to said request, looking in said first cache for particular metadata associated with said file; and

if said first cache does not have said particular metadata associated with said file, then looking in a third cache for said particular metadata associated with said file,

wherein said third cache only contains file metadata that has been committed by the database system.

70. The computer-readable medium of claim 69 further comprising instructions for performing the step of sending a request for said particular metadata to said database system if said particular metadata is not in said third cache.

71. A method for managing a cache associated with a repository, the method comprising the steps of:

receiving, at said cache, a request for a piece of information;

determining whether the piece of information is stored in said cache;

if the information is not stored in said cache, accessing said repository to determine whether the piece of information is stored in said repository;

if the information is not stored in said repository, then inserting into said cache a negative cache entry that identifies said piece of information and indicates that said piece of information does not reside in said repository;

in response to a subsequent request for said piece of information, determining that said piece of information does not reside in said repository based on said negative cache entry, without involving said repository in handling of said subsequent request.

72. The method of claim 71 further comprising the steps of:

detecting occurrence of an action that may cause said repository to contain said piece of information; and

invalidating said negative cache entry in response to detecting said action.

73. The method of claim 71 wherein:

the cache is external to a database server that manages a database that contains information about files; and

the step of inserting into said cache a negative cache entry includes inserting into said cache a negative cache entry that indicates that the database does not contain certain metadata information about one or more of said files.

74. The method of claim 73 wherein the step of inserting a negative cache entry includes inserting a negative cache entry that indicates that a certain folder does not exist at a particular location in a folder hierarchy.

75. The method of claim 73 wherein the step of inserting a negative cache entry includes inserting a negative cache entry that indicates that the database does not contain file system metadata about a certain file.

76. The method of claim 71 wherein the cache includes both

positive cache entries that indicate that the repository contains certain pieces of information, and

negative cache entries that indicate that the repository does not contain certain pieces of information.