Addressing the name space mismatch between content servers and content caching systems

• US 6,985,936 B2
• Filed: 09/27/2001
• Issued: 01/10/2006
• Est. Priority Date: 09/27/2001
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Doyle, Ronald P., Agarwalla, Rajesh S., Jiang, Tianyu, Niranjan, Thirumale, Ramamurthy, Srikanth
• Primary Examiner(s): Wiley, David
• Assistant Examiner(s): NEURAUTER, GEORGE C
• Application Number:
  US09/966,229
• Publication Number:
  US 20030061278A1
• Time in Patent Office: 1,566 Days
• Field of Search: 709/203, 709/213, 709/219, 709/221, 709/223, 709/222, 707/10, 707/100, 711/141
• US Class Current: 709/221
• CPC Class Codes: G06F 16/9574 of access to content, e.g. by caching
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/561 Adding application-functional data or data for application control, e.g. adding metadata
• CPC Class Codes: H04L 67/564 Enhancement of application control based on intercepted application data
• CPC Class Codes: H04L 67/5682 Policies or rules for updating, deleting or replacing the stored data
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]

Abstract

Techniques are disclosed for addressing the name space mismatch between content caching systems (which use Uniform Resource Locators, or “URLs”) and content servers (which use file and path names). A file name-to-URL mapping is created for use by content caching systems, and data in protocol response messages (and optionally in protocol request messages) is augmented to transmit information for use in creating this mapping, enabling a content caching system to automatically and dynamically populate its file name-to-URL mapping. By having the file name available, the caching system can now respond to content management messages which identify the cached content by only the content'"'"'s associated file name. Techniques for encoding the message extensions include: use of new directives on existing cache-control headers in Hypertext Transfer Protocol (“HTTP”) messages; addition of new headers in HTTP messages; and use of meta-data in markup languages such as Hypertext Markup Language (“HTML”) or Extensible Markup Language (“XML”) format.

Claims

1. A method of addressing a name space mismatch between content servers and content caching systems, comprising steps of:

caching one or more content elements in a content caching system, wherein the cached elements are identified by the content caching system using a uniform resource locator format;

automatically learning mappings between a directory structure format and the uniform resource locator format, responsive to receiving the content elements for caching, wherein the received content elements are identifiable to the content caching system using the uniform resource locator format but also have associated therewith identifiers in the directory structure format, the mapping for each received content element storing its identifier in the uniform resource locator format and its associated identifier in the directory structure format; and

using the mappings when updating one or more selected cached content elements, responsive to receiving notifications of changed content elements that identify the changed content elements using the directory structure format, by requesting, by the content caching system, updated versions of the selected cached content elements from a content server, wherein each request identifies one of the selected cached content elements using the identifier, in the uniform resource locator format, from the mapping that stores the identifier, in the directory structure format, from the received notification for that content element.

2. The method according to claim 1, further comprising the step of using the mappings to invalidate one or more particular cached content elements by invoking an invalidation function for the content caching system, by the content caching system responsive to receiving notifications of invalid content elements that identify the particular content elements using the directory structure format, wherein each invocation identifies one of the particular cached content elements using the identifier, in the uniform resource locator format, from the mapping that stores the identifier, in the directory structure format, from the received notification for that content element.

3. A method of addressing a name space in mismatch between content servers and content caching systems, comprising steps of:

automatically learning mappings between a directory structure format which identifies content elements accessible from one or more content servers and a uniform resource locator format which identifies content elements cached by a caching system, when the caching system receives the content elements from the content servers for caching, wherein the received content elements are identifiable to the caching system using the uniform resource locator format and also have associated therewith corresponding identifiers in the directory structure format, the mapping for each received content element storing its identifier in the uniform resource locator format and its corresponding identifier in the directory structure format;

receiving, at the caching system, a notification pertaining to one of the cached content elements, wherein the notification identifies the content element using its identifier in the directory structure format;

consulting the learned mappings using the identifier in the directory structure format from the notification, thereby determining the corresponding identifier in the uniform resource locator format; and

processing the received notification using the corresponding identification in the uniform resource locator format.

4. The method according to claim 3, wherein:

the identifier in the directory structure format identifies more than one of the cached content elements;

the consulting step determines corresponding identifiers, in the uniform resource locator format, of the more than one identified cached content elements; and

the processing step processes each of the identified cached content elements using the corresponding identifiers in the uniform resource locator format.

5. The method according to claim 3, wherein the notification is a content update notification.

6. The method according to claim 3, wherein the notification is a content invalidation notification.

7. The method according to claim 3, wherein the notification is received from a content management system.

8. The method according to claim 3, wherein a plurality of notifications are received, the plurality comprising at least one content update notification and at least one content invalidation notification, and wherein the plurality of notifications use the directory structure format for conveying the notifications to the content caching system and also to the content servers.

9. The method according to claim 3, wherein the automatically learning step operates when the content elements are received during cache miss processing.

10. The method according to claim 3, wherein the corresponding identifier directory structure format is specified, for each of the received content elements, in a response message with which the content element is received.

11. A method of addressing a name space mismatch between content caching systems and content servers, comprising steps of:

sending a content request from a content caching system to a content server, wherein the content request identifies requested content using a first identifier in a uniform resource locator format;

receiving a content response at the content caching system from the content server, wherein the content response provides the requested content, identifiable using the first identifier, and is augmented with a second identifier of the requested content in a directory structure format; and

creating an entry in a mapping, responsive to the receiving step, wherein the entry maps the second identifier in the directory structure format to the first identifier in the uniform resource locator format.

12. The method according to claim 11, wherein the content request signifies an ability to process the augmented content response.

13. The method according to claim 12, wherein the ability is signified using a message header of the content request.

14. The method according to claim 13, wherein the message header specifies that the caching system is content distribution aware.

15. The method according to claim 13, wherein the message header is a Hypertext Transfer Protocol (“

HTTP”

) header.

16. The method according to claim 12, wherein the ability is signified using syntax of a markup language.

17. The method according to claim 16, wherein the markup language is Hypertext Markup Language (“

HTML”

).

18. The method according to claim 17, wherein the syntax comprises a “

META”

tag using an “

HTTP-EQUIV”

attribute syntax.

19. The method according to claim 17, wherein the syntax comprises a “

META”

tag using a “

NAME”

attribute syntax.

20. The method according to claim 17, wherein the syntax comprises a specially-denoted comment.

21. The method according to claim 16, wherein the markup language is XML (“

Extensible Markup Language”

).

22. The method according to claim 12, wherein the ability is signified using a cookie on the content request.

23. The method according to claim 11, wherein the augmented content response provides the second identifier using a header.

24. The method according to claim 23, wherein the header comprises an extended cache control header.

25. The method according to claim 23, wherein the header comprises a file identification header.

26. The method according to claim 25, wherein the file identification header is a Hypertext Transfer Protocol (“

HTTP”

) header.

27. The method according to claim 11, wherein the augmented content response provides the second identifier using syntax of a markup language.

28. The method according to claim 27, wherein the markup language is Hypertext Markup Language (“

HTML”

).

29. The method according to claim 28, wherein the syntax comprises a “

META”

tag using an “

HTTP-EQUIV”

attribute syntax.

30. The method according to claim 28, wherein the syntax comprises a “

META”

tag using a “

NAME”

attribute syntax.

31. The method according to claim 28, wherein the syntax comprises a specially-denoted comment.

32. The method according to claim 27, wherein the markup language is XML (“

Extensible Markup Language”

).

33. The method according to claim 11, wherein the augmented content response provides the second identifier using a cookie on the content request.

34. The method according to claim 11, further comprising the step of caching the received content at the content caching system, responsive to the receiving step.

35. The method according to claim 34, further comprising the steps of:

receiving an update notification for a selected previously-cached content at the content caching system, wherein the update notification identifies the selected previously-cached content using a selected identifier thereof in the directory structure format;

consulting the mapping, using the selected identifier, to retrieve the first identifier in the uniform resource format; and

using the retrieved first identifier to request an update of the selected previously-cached content.

36. The method according to claim 35, wherein the using step further comprises steps of:

sending a new content request to the content server, wherein the new content request identifies the requested content using the retrieved first identifier;

receiving a new content response from the content server, wherein the new content response provides the requested updated content; and

caching the received updated content.

37. The method according to claim 34, further comprising the steps of:

receiving an invalidation notification for a selected previously-cached content at the content caching system, wherein the invalidation notification identifies the selected previously-cached content using a selected identifier in the directory structure format;

consulting the mapping, using the selected identifier, to retrieve the first identifier in the uniform resource locator format; and

using the retrieved first identifier to request an invalidation of the selected previously-cached content.

38. The method according to claim 11, wherein the second identifier is a file path and a file name.

39. The method according to claim 11, wherein the second identifier is a substituted file path and file name.

40. The method according to claim 39, wherein the substituted file path and file name is an encrypted file path and file name.

41. The method according to claim 11, further comprising steps of:

receiving, at the content server, the sent content request;

obtaining, by the content server, the requested content;

preparing a response message, by the content server, to transmit the obtained content, wherein the response message is augmented with the second identifier in the directory structure format; and

sending the prepared response message, including the obtained content, from the content server to the content caching system.

42. A system for addressing a name space mismatch between content servers and content caching systems, comprising:

means for automatically learning mappings between a directory structure format which identifies content elements accessible from one or more content servers and a uniform resource locator format which identifies content elements cached by a caching system, when the caching system receives the content elements from the content servers for caching, wherein the received content elements are identifiable to the caching system using the uniform resource locator format and also have associated therewith corresponding identifiers in the directory structure format, the mapping for each received content element storing its identifier in the uniform resource locator format and its corresponding identifier in the directory structure format;

means for receiving, at the caching system, a notification pertaining to one of the cached content elements, wherein the notification identifies the content clement using its identifier in the directory structure format;

means for consulting the learned mappings using the identifier in the directory structure format from the notification, thereby determining the corresponding identifier in uniform resource locator format; and

means for processing the received notification using the corresponding identification in the uniform resource locator format.

43. A system for addressing a name space mismatch between content caching systems end content servers, comprising:

means for sending a content request from a content caching system to a content server, wherein the content request identifies requested content using a first identifier in a uniform resource locator format;

means for receiving a content response at the content caching system from the content server, wherein the content response provides the requested content, identifiable using the first identifier, and is augmented with a second identifier of the requested content in a directory structure format; and

means for creating an entry in a mapping, responsive to the means for receiving, wherein the entry maps the second identifier in the directory structure format to the first identifier in the uniform resource locator format.

44. A computer program product for addressing a name space mismatch between content sewers and content caching systems, the computer program product embodied on one &

mare computer-usable media and comprising;

computer readable program code means for automatically learning mappings between a directory structure format which identifies content elements accessible from one or more content servers and a uniform resource locator format which identifies content elements cached by a caching system, when the caching system receives the content elements from the content servers for caching, wherein the received content elements are identifiable to the caching system using the uniform resource locator format and also have associated therewith corresponding identifiers in the directory structure format, the mapping for each received content element storing its identifier in the uniform resource locator format and its corresponding identifier in the directory structure format;

computer readable program code means for receiving, at the caching system, a notification pertaining to one of the cached content elements, wherein the notification identifies the content element using its identifier in the directory structure format;

computer readable program code means for consulting the learned mappings using the identifier in the directory structure format from the notification, thereby determining the corresponding identifier in the uniform resource locator format; and

computer readable program code means for processing the received notification using the corresponding identification in the uniform resource locator format.

45. A computer program product for addressing a name space mismatch between content servers and content caching systems, the computer program product embodied on one or more computer-usable media and comprising, comprising:

computer readable program code means for sending a content request from a content caching system to a content sewer, wherein the content request identifies requested content using a first identifier in a uniform resource locator format;

computer readable program code means for receiving a content response at the content caching system from the content server, wherein the content response provides the requested content, identifiable using the first identifier, and is augmented with a second identifier of the requested content a directory structure format; and

computer readable program code means for creating an entry in a mapping, responsive to the computer readable program code means for receiving, wherein the entry maps the second identifier in the directory structure format to the first identifier in the uniform resource locator format.