Field level concurrency and transaction control for out-of-process object caching

• US 8,364,903 B2
• Filed: 12/22/2009
• Issued: 01/29/2013
• Est. Priority Date: 12/22/2009
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Verizon Patent and Licensing Incorporated (Verizon Communications Inc.)
• Inventors: Singh, Amit, Jindal, Tarun, Yang, Tao, Ebrahimi, Fairborz, Kavouspour, Dinyar
• Primary Examiner(s): Shin, Christopher B
• Application Number:
  US12/644,125
• Publication Number:
  US 20110154315A1
• Time in Patent Office: 1,134 Days
• Field of Search: 711/118, 711141-143, 711/147, 717/171, 709/213
• US Class Current: 711/141
• CPC Class Codes: G06F 9/526 Mutual exclusion algorithms

Abstract

A method includes executing a multi-threaded, object-oriented application (OOA) on a device; receiving, by multiple threads of the OOA, an object from an out-of-process cache memory; mutating one or more fields of the object, wherein the one or more fields correspond to one or more attributes of the object; and applying an update of the one or more fields that have been mutated to the out-of-process cache memory, wherein the applying the update updates the one or more fields mutated at a field level and not at an object level.

Claims

1. A method comprising:

executing a multi-threaded, object-oriented application (OOA) on a device;

receiving, by multiple threads of the OOA, an object from an out-of-process cache memory;

mutating one or more fields of the object, wherein the one or more fields correspond to one or more attributes of the object; and

applying an update of the one or more fields that have been mutated to the out-of-process cache memory, wherein the applying the update updates the one or more fields mutated at a field level and not at an object level.

2. The method of claim 1, wherein the one or more fields are disjoint with respect to each thread.

3. The method of claim 1, further comprising:

collecting the one or more fields of the object as the one or more fields are being mutated.

4. The method of claim 1, further comprising:

determining when the one or more fields mutated are to be updated to the out-of-process cache.

5. The method of claim 4, wherein the determining is based on the multi-threaded OOA'"'"'s visibility of the object or a thread notification boundary, wherein the object corresponds to a shared object.

6. The method of claim 1, wherein the multi-threaded OOA includes a JAVA application.

7. The method of claim 1, wherein the mutating comprises:

receiving at least one of a user request to cause the one or more fields to be mutated or a non-user request to cause the one or more fields to be mutated.

8. The method of claim 7, wherein the non-user request comprises a daemon process.

9. A device comprising:

one or more memories to store instructions; and

one or more processing systems to execute the instructions in the one or more memories to;

execute a multi-threaded, object-oriented application (OOA);

obtain a shared object by multiple threads of the OOA;

mutate one or more fields of the shared object; and

apply an update of the one or more fields mutated to an out-of-process cache memory, wherein the applying updates the one or more fields mutated at a field level and not at an object level.

10. The device of claim 9, wherein the device comprises an application server.

11. The device of claim 9, wherein the one or more fields are disjoint with respect to each of the multiple threads.

12. The device of claim 9, wherein the one or more processing systems execute the instructions to:

determine when the one or more fields of the shared object are to be updated to the out-of-process cache, wherein the determining is based on a visibility of the shared object associated with at least one of the multiple threads.

13. The device of claim 9, wherein the one or more processing systems execute the instructions to:

collect the one or more fields of the shared object as the one or more fields are being mutated, wherein the one or more fields are disjoint; and

issue a command to a coherence client to apply the update of the one or more fields collected to the out-of-process cache.

14. The device of claim 9, wherein the one or more processing systems execute the instructions to:

issue a command to a coherence client to apply the update of the one or more fields when the shared object goes outside a scope of a method call associated with at least one of the multiple threads.

15. The device of claim 9, wherein the multi-threaded OOA comprises one of a JAVA application, a C++ application, or a .NET application.

16. One or more computer-readable media containing instructions executable by at least one processor, the one or more computer-readable media storing instructions to:

execute a multi-threaded, object-oriented application (OOA);

obtain a shared object by multiple threads of the OOA;

mutate one or more fields of the shared object;

determine when the one or more fields are to be updated to an out-of-process cache memory; and

apply an update of the one or more fields mutated to the out-of-process cache memory, wherein the applying updates the one or more fields mutated at a field level and not at an object level.

17. The one or more computer-readable media of claim 16, wherein the one or more fields are disjoint fields.

18. The one or more computer-readable media of claim 16, wherein the multi-threaded OOA comprises one of a JAVA application, a C++ application, or a .NET application.

19. The one or more computer-readable media of claim 16, further storing one or more instructions to:

collect the one or more fields of the shared object as the one or more fields are being mutated; and

issue a command to apply the update in a batch fashion.

20. The one or more computer-readable media of claim 16, wherein the instructions to determine comprise one or more instructions to:

determine based on a visibility of the shared object associated with at least one of the multiple threads.