CACHE REGION CONCEPT

• US 20100268881A1
• Filed: 07/07/2010
• Published: 10/21/2010
• Est. Priority Date: 12/28/2004
• Status: Active Grant

• Current Assignee: SAP SE
• Original Assignee: Dirk Marwinski, Frank Kilian, Galin Galchev, Oliver Luik, Petio G. Petev
• Inventors: Kilian, Frank, Galchev, Galin, Luik, Oliver, Marwinski, Dirk, Petev, Petio G.
• Granted Patent: US 7,996,615 B2
• Time in Patent Office: Days
• US Class Current: 711/121
• CPC Class Codes: G06F 12/0253 Garbage collection, i.e. reclamation of unreferenced memory
• CPC Class Codes: G06F 12/084 with a shared cache
• CPC Class Codes: G06F 12/0842 for multiprocessing or multitasking
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 12/121 using replacement algorithms
• CPC Class Codes: G06F 2212/1021 Hit rate improvement
• CPC Class Codes: G06F 2212/282 Partitioned cache
• CPC Class Codes: G06F 9/455 Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines

Abstract

A method to associate a storage policy with a cache region is disclosed. In this method, a cache region associated with an application is created. The application runs on virtual machines, and where a first virtual machine has a local memory cache that is private to the first virtual machine. The first virtual machine additionally has a shared memory cache that is shared by the first virtual machine and a second virtual machine. Additionally, the cache region is associated with a storage policy. Here, the storage policy specifies that a first copy of an object to be stored in the cache region is to be stored in the local memory cache and that a second copy of the object to be stored in the cache region is to be stored in the shared memory cache.

Claims

1. A system comprising:

a cache manager to;

create a first cache region associated with an application, the application running on a plurality of virtual machines, a first one of the plurality of virtual machines having a local memory cache that is private to the first one of the plurality of virtual machines and a shared memory cache that is shared by the first one of the plurality of virtual machines and a second one of the plurality of virtual machines; and

associate the first cache region with a first storage policy, the first storage policy specifying an object to be stored in the first cache region is to be stored in the local memory cache and that a garbage collector is not to remove the object from the local memory cache.

2. The system of claim 1, wherein the first storage policy further specifies that an eviction of the object from the first cache region is to result in a storage of the object in a storage resource having a higher latency than the local memory cache.

3. The system of claim 2, wherein the storage resource is a local memory associated with the local memory cache.

4. The system of claim 1, wherein the cache manager is further to:

receive a request from the application to store the object in the first cache region;

in response to the receiving of the request from the application to store the object, store the object in the local memory cache and establish a hard reference to the object, an identifying of the hard reference by the garbage collector to prevent the garbage collector from removing the object from the local memory cache regardless of a lack of a use of the object by the application.

5. The system of claim 1, wherein the cache manager is further to:

create a second cache region associated with the application; and

associate a second storage policy with the second storage region, the second storage policy specifying that an additional object to be stored in the second cache region is to be stored in the local memory cache and that a lack of a use of the additional object by the application is to enable the garbage collector to remove the additional object from the local memory cache.

6. The system of claim 5, wherein the cache manager is further to:

receive a request from the application to store the additional object in the second cache region; and

in response to the receiving of the request from the application to store the additional object, store the additional object in the local memory cache and establish a soft reference to the additional object, an identifying of the soft reference by the garbage collector enabling the garbage collector to remove the additional object from the local memory cache unless the application has a hard reference to the additional object, an identifying of the hard reference by the garbage collector preventing the garbage collector from removing the additional object from the local memory cache.

7. A method comprising:

creating a first cache region associated with an application, the application running on a plurality of virtual machines, a first one of the plurality of virtual machines having a local memory cache that is private to the first one of the plurality of virtual machines and a shared memory cache that is shared by the first one of the plurality of virtual machines and a second one of the plurality of virtual machines; and

associating the first cache region with a first storage policy, the first storage policy specifying that a first copy of an object to be stored in the first-cache region is to be stored in the local memory cache and that a second copy of the object to be stored in the first cache region is to be stored in the shared memory cache.

8. The method of claim 7, further comprising:

receiving a request from the application to retrieve the object, the object having been stored in the first cache region; and

in response to the receiving of the request from the application to retrieve the object, detecting that a second version of the object is newer than a first version of the object, the second version of the object being stored in the shared memory cache and the first version of the object being stored in the local memory cache; and

in response to the detecting, providing the second version of the object to the application.

9. The method of claim 8, wherein the providing of the second version of the object comprises copying the second version of the object from the shared memory cache to the local memory cache.

10. The method of claim 9, wherein the providing of the second version of the object further comprises enabling the application to establish a hard reference to the second version of the object in the local memory cache, an identifying of the hard reference by a garbage collector preventing the garbage collector from removing the second version of the object from the local memory cache.

11. The method of claim 7, wherein the first storage policy further specifies that an eviction of the object from the local memory cache is to result in a storage of the object in a storage resource having a higher latency than the local memory cache.

12. The method of claim 11, wherein the storage resource is a shared memory associated with the shared memory cache.

13. The method of claim 7, further comprising:

creating a second cache region associated with the application; and

associating the second cache region with a second storage policy, the second storage policy specifying that a first version of an additional object to be stored in the second cache region is to be stored in the shared memory cache and that the first version of the additional object is not to be modified.

14. The method of claim 13, further comprising:

receiving a request from the application to incorporate a modification into the additional object, the additional object having been stored in the second cache region; and

in response to the receiving of the request from the application to incorporate the modification, incorporating the modification into a second version of the additional object and storing the second version of the additional object in the shared memory cache.

15. A machine-readable storage medium comprising instructions that, when executed by a processor of a machine, causes the machine to perform a method comprising:

creating a cache region associated with an application, the application running on a plurality of virtual machines, a first one of the plurality of virtual machines having a local memory cache that is private to the first one of the plurality of virtual machines and a shared memory cache that is shared by the first one of the plurality of virtual machines and a second one of the plurality of virtual machines; and

associating the cache region with a storage policy, the storage policy specifying that a first shared closure stored in the cache region is to be bundled with a second shared closure stored in the cache region based on a size of the first shared closure and a size of the second shared closure not transgressing a size threshold, the first shared closure being a first plurality of objects, each of the first plurality of objects shareable by the plurality of virtual machines and not referencing an object outside the first plurality of objects, the second shared closure being a second plurality of objects, each of the second plurality of objects shareable by the plurality of virtual machines and not referencing an object outside the second plurality of objects.

16. The machine-readable storage medium of claim 15, the method further comprising:

receiving a request from the application to retrieve the first shared closure from the cache region;

in response to the receiving of the request from the application to retrieve the first shared closure, detecting that the first shared closure is bundled with the second shared closure; and

in response to the detecting, providing the application with the first shared closure and the second shared closure.

17. The machine-readable storage medium of claim 16, wherein the providing of the application with the first shared closure and the second shared closure comprises transporting the first shared closure and the second shared closure from the shared memory cache to the local memory cache using a single transfer process.

18. The machine-readable storage medium of claim 15, wherein the bundling of the first shared closure with the second shared closure is also based on a determination that a number of shared closures transported between the shared memory cache and the local memory cache within a time period transgresses a quantity threshold.

19. The machine-readable storage medium of claim 15, the method further comprising associating the cache region with an eviction policy, the eviction policy specifying that an object stored in the cache region is to be evicted based on an attribute of the object regardless of a position of the object in a sorting queue associated with the eviction policy.

20. The machine-readable storage medium of claim 15, wherein the attribute is one of a time-to-live attribute and an absolute eviction time attribute.