Adjusting the cacheability of web documents according to the responsiveness of its content server

• US 6,823,374 B2
• Filed: 11/04/2002
• Issued: 11/23/2004
• Est. Priority Date: 11/16/2001
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Fineground Networks LLC
• Inventors: Race, Daren A., Kausik, Balas Natarajan, Jawahar, Janardhanan
• Primary Examiner(s): LIN, WEN TAI
• Application Number:
  US10/287,364
• Publication Number:
  US 20030101231A1
• Time in Patent Office: 750 Days
• Field of Search: 711/124, 711/135, 711/165, 709/203, 709/213, 709/217, 709/218, 709/219
• US Class Current: 709/219
• CPC Class Codes: H04L 67/02 based on web technology, e.g. hypertext transfer protocol [HTTP]
• CPC Class Codes: H04L 67/563 Data redirection of data network streams
• 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]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: H05K 3/285 Permanent coating compositions

Abstract

We disclose techniques for varying the caching of content provided by a content server as a function of the server'"'"'s load. When the server is lightly loaded, freshness of the content is maintained. As server load increases, caching time increases, to trade off freshness against transmission time. Similarly, when the server is heavily loaded, users might quickly be served cached content that is only slightly stale—as opposed to content that was fresh at the time of the request, but which becomes materially stale by the time the server completes serving the response. The server'"'"'s load can be measured by its response time to a request, or otherwise. Optionally, the system can override the load-based caching by defining classes of requests that automatically trigger updating or refreshing (e.g., cache expiration) of related information in the cache.

Claims

1. A method for varying the cacheability of networked content depending on the responsiveness of a content server, comprising:

(a) computing a measure of responsiveness of a content server;

(b) determining a caching interval for a requested item in accordance with said responsiveness, including;

(i) setting the caching interval lower if an actual level of responsiveness is higher than a responsiveness threshold; and

(ii) else setting the caching interval higher;

(c) obtaining said requested item from said content server; and

(d) specifying a cacheability of said item in accordance with said determined caching interval.

2. The method of claim 1 where:

(i) said content servers responsiveness is based on an average response time to requests for content; and

(ii) said responsiveness threshold is based on a desired response time.

3. The method of claim 1 where:

(i) said content servers responsiveness is based on an actual response time to requests for content; and

(ii) said responsiveness threshold is based on a desired response time.

4. The method of claim 1 where content server'"'"'s responsiveness is based on a number of pages served by said content server in a specified time.

5. The method of claim 1 where content server'"'"'s responsiveness is based on said content server'"'"'s CPU load.

6. The method of claim 1 where content server'"'"'s responsiveness is based on an overall demand being imposed by active users of said content server.

7. The method of claim 1 where said caching interval represents a difference between said computed responsiveness and said responsiveness threshold.

8. The method of claim 1 wherein said (c) occurs before said (b).

9. The method of claim 1 where said responsiveness threshold is determined based on a statistical measure of historical requests.

10. The method of claim 1 where said (d) includes caching said item if said computed responsiveness is worse than said responsiveness threshold.

11. The method of claim 1 where said (d) includes not caching said item if said computed responsiveness is better than said responsiveness threshold.

12. The method of claim 1 where said (d) includes not caching said item if an aspect of said requested item would trigger invalidity of said cached item.

13. The method of claim 1 where an aspect of said requested item includes information in an identifier of said item.

14. The method of claim 1 where said aspect of a requested item includes information embedded within said item.

15. The method of claim 1 further comprising, after at least said (b), expiring at least one cached item in response to a determination that information in said requested item would trigger invalidity of an aspect of said cached item.

16. The method of claim 1 implemented at a proxy server between said content server and a user.

17. A method for obtaining networked content from a cache or a content server, depending on the responsiveness of the content server, comprising:

(a) receiving a users request for an item;

(b) computing a measure of responsiveness of a content server;

(c) obtaining said requested item;

(i) from said content server if said computed responsiveness is higher than a responsiveness threshold;

else

(ii) from a cache containing a previously cached version of said item;

(iii) setting a caching interval lower if an actual level of responsiveness is higher than a responsiveness threshold.

18. The method of claim 17 where said request triggers invalidation of at least some of the cached items in the cache.

19. The method of claim 17 where said (c) includes extending a cache life of said previously cached version while said computed responsiveness is higher than a responsiveness threshold.

20. A method for obtaining networked content from a cache or a content server, depending on the responsiveness of the content server, comprising:

(a) receiving a user request for an item available at a content server;

(b) determining whether said user request will trigger invalidity of a previously cached item;

(c) deleting from a cache all items determined to be invalid as in (b);

(i) setting a caching interval lower if an actual level of responsiveness is higher than a responsiveness threshold.

21. A computer-readable medium comprising logic instructions for varying the cacheability of networked content depending on the responsiveness of a content server, said instructions when executed on a computer:

(a) computing a measure of responsiveness of a content server;

(b) determining a caching interval for a requested item in accordance with said responsiveness, including;

(i) setting the caching interval lower if an actual level of responsiveness is higher than a responsiveness threshold;

else

(ii) setting the caching interval higher;

(c) obtaining said requested item from said content server; and

(d) specifying a cacheability of said item in accordance with said determined caching interval.

22. The computer-readable medium of claim 21 where said caching interval represents a difference between said computed responsiveness and said responsiveness threshold.

23. The computer-readable medium of claim 21 where said (d) includes caching said item if said computed responsiveness is worse than said responsiveness threshold.

24. The computer-readable medium of claim 21 where said (d) includes not caching said item if said computed responsiveness is better than said responsiveness threshold.

25. The computer-readable medium of claim 21 where said (d) includes not caching said item if an aspect of said requested item would trigger invalidity of said cached item.

26. The computer-readable medium of claim 21 further comprising logic instructions for expiring at least one cached item in response to a determination that information in said requested item would trigger invalidity of an aspect of said cached item.

27. The computer-readable medium of claim 21 implemented at a proxy server between said content server and a user.

28. A device for varying the cacheability of networked content depending on the responsiveness of a content server, comprising:

(a) means for computing a measure of responsiveness of a content server;

(b) means for determining a caching interval for a requested item if said computed measure of responsiveness is higher than a responsiveness threshold and else setting the caching interval higher;

(c) an interface for obtaining said requested item from said content server; and

(d) means for specifying a cacheability of said item in accordance with said determined caching interval.

29. A proxy server for varying the cacheability of networked content depending on the responsiveness of a content server, comprising:

(a) an interface to a content server;

(b) computer-executable program logic for determining a caching interval for a requested item in accordance with a responsiveness of said content server, including;

(i) setting the caching interval lower an actual level of responsiveness is higher than a responsiveness threshold; and

(ii) setting the caching interval higher if an actual level of responsiveness is higher than said responsiveness threshold; and

(c) computer-executable program logic for specifying a cacheability of said item in accordance with said determined caching interval.

30. The proxy server of claim 29 where said (d) includes not caching said item if an aspect of said requested item would trigger invalidity of said cached item.

31. The proxy server of claim 29 further comprising computer-executable program logic for expiring at least one cached item in response to a determination that information in said requested item would trigger invalidity of an aspect of said cached item.