System and Method for Caching Client Requests to an Application Server Based on the Application Server's Reliability

• US 20080126831A1
• Filed: 07/27/2006
• Published: 05/29/2008
• Est. Priority Date: 07/27/2006
• Status: Abandoned Application

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Venkataraman, Sundararaman, Downey, Audra F., Peters, Mark E., Subramanian, Balan, Sundararaman, Renganathan
• Application Number:
  US11/460,289
• Publication Number:
  US 20080126831A1
• Time in Patent Office: Days
• US Class Current: 714/4
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: H04L 67/56 Provisioning of proxy services store-and-forward switching systems in data switching networks H04L12/54
• CPC Class Codes: H04L 67/5683 Storage of data provided by user terminals, i.e. reverse caching
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

An Intelligent Caching Tool collects reliability statistics for an application server to build a Hidden Markov Model. Using the Hidden Markov Model, the Intelligent Caching Tool calculates a reliability index for the application server. After setting a user defined reliability threshold, the Intelligent Caching Tool caches all client requests and the status of the application server'"'"'s response when the reliability index is below the reliability threshold.

Claims

1. A computer implemented process for caching client requests based on the predicted reliability of an application server, the computer implemented process comprising:

collecting reliability statistics for the application server;

building a Hidden Markov Model using the reliability statistics;

calculating a reliability index for the application server;

setting an upper reliability threshold; and

saving all client requests and the status of the application server'"'"'s responses to a cache if the reliability index is below the upper reliability threshold.

2. The computer implemented process of claim 1 further comprising using the cache to fail-over client/server sessions if the application server has a failure.

3. The computer implemented process of claim 1 further comprising deleting the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the client/server session ends.

4. The computer implemented process of claim 1 further comprising identifying at least one commit point when the application server response includes streaming media and deleting the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the at least one commit point is reached.

5. The computer implemented process of claim 1 further comprising:

setting a lower reliability threshold; and

setting a cache size limit for the cache of application servers with a reliability index between the upper reliability threshold and the lower reliability threshold.

6. The computer implemented process of claim 5 further comprising deleting the oldest client request and the status of the application server'"'"'s responses from the cache when the cache size exceeds the cache size limit.

7. The computer implemented process of claim 1 wherein the reliability index calculation includes predictive failures based on known risk factors.

8. An apparatus for caching client requests based on the predicted reliability of an application server, the apparatus comprising:

a processor;

a memory connected to the processor;

an application running in the memory accessible by a remote client;

an intelligent caching tool program in the memory operable to collect reliability statistics for the application server, build a Hidden Markov Model using the reliability statistics, calculate a reliability index for the application server, set an upper reliability threshold, and save all client requests and the status of the application server'"'"'s responses to a cache if the reliability index is below the upper reliability threshold.

9. The apparatus of claim 8 wherein the intelligent caching tool program in the memory is further operable to use the cache to fail-over client/server sessions if the application server has a failure.

10. The apparatus of claim 8 wherein the intelligent caching tool program in the memory is further operable to delete the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the client/server session ends.

11. The apparatus of claim 8 wherein the intelligent caching tool program in the memory is further operable to identify at least one commit point when the application server response includes streaming media and delete the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the at least one commit point is reached.

12. The apparatus of claim 8 wherein the intelligent caching tool program in the memory is further operable to set a lower reliability threshold and set a cache size limit for the cache of application servers with a reliability index between the upper reliability threshold and the lower reliability threshold.

13. The apparatus of claim 12 wherein the intelligent caching tool program in the memory is further operable to delete the oldest client request and the status of the application server'"'"'s responses from the cache when the cache'"'"'s size exceeds the cache size limit.

14. A computer readable memory containing a plurality of executable instructions to cause a computer to cache client requests based on the predicted reliability of an application server, the plurality of instructions comprising:

a first instruction to collect reliability statistics for the application server;

a second instruction to build a Hidden Markov Model using the reliability statistics;

a third instruction to calculate a reliability index for the application server;

a fourth instruction to send setting an upper reliability threshold; and

a fifth instruction to save all client requests and the status of the application server'"'"'s responses to a cache if the reliability index is below the upper reliability threshold.

15. The computer readable memory of claim 14 further comprising an instruction to using the cache to fail-over client/server sessions if the application server has a failure.

16. The computer readable memory of claim 14 further comprising an instruction to delete the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the client/server session ends.

17. The computer readable memory of claim 14 further comprising an instruction to identify at least one commit point when the application server response includes streaming media and delete the client requests and status of the application server'"'"'s responses from the cache for a client/server session once the at least one commit point is reached.

18. The computer readable memory of claim 14 further comprising:

an instruction to set a lower reliability threshold; and

an additional instruction to set a cache size limit for the cache of application servers with a reliability index between the upper reliability threshold and the lower reliability threshold.

19. The computer readable memory of claim 18 further comprising an instruction to delete the oldest client request and the status of the application server'"'"'s responses from the cache when the cache'"'"'s size exceeds the cache size limit.

20. The computer readable memory of claim 14 wherein the third instruction to calculate a reliability index includes calculating predictive failures based on known risk factors.