Data storage management using a distributed cache scheme

• US 8,782,323 B2
• Filed: 10/30/2009
• Issued: 07/15/2014
• Est. Priority Date: 10/30/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Glikson, Alex, Goikhman, Shay, Rochwerger, Benny
• Primary Examiner(s): Bradley, Matthew
• Assistant Examiner(s): GOLDSCHMIDT, CRAIG S
• Application Number:
  US12/609,023
• Publication Number:
  US 20110138102A1
• Time in Patent Office: 1,719 Days
• Field of Search: 711/6
• US Class Current: 711/6
• CPC Class Codes: G06F 12/0813 with a network or matrix configuration
• CPC Class Codes: G06F 12/0815 Cache consistency protocols
• CPC Class Codes: G06F 9/45533 Hypervisors; Virtual machine monitors

Abstract

A method for accessing data stored in a distributed storage system is provided. The method comprises determining whether a copy of first data is stored in a distributed cache system, where data in the distributed cache system is stored in free storage space of the distributed storage system; accessing the copy of the first data from the distributed cache system if the copy of the first data is stored in a first data storage medium at a first computing system in a network; and requesting a second computing system in the network to access the copy of the first data from the distributed cache system if the copy of the first data is stored in a second data storage medium at the second computing system. If the copy of the first data is not stored in the distributed cache system, the first data is accessed from the distributed storage system.

Claims

1. A computer-implemented method for accessing data stored in a distributed storage system, the method comprising:

maintaining a distributed storage system comprising a plurality of virtual machines executed on a plurality of computing systems connected over a network, wherein a portion of a respective data storage volume on each of the plurality of computing systems is allocated to one of the plurality of virtual machines executed on the respective computing system as a respective virtual memory,

wherein upon addition of each of the plurality of computing systems to the network, determining whether there is sufficient free storage space on the respective computing system to be allocated to a distributed cache system implemented over the distributed storage system, the plurality of computing systems sharing free space in the virtual memories allocated to the plurality of virtual machines based on cache metadata identifying the amount of free storage space available on one or more of said plurality of computing systems;

determining, based on metadata associated with first data, whether a copy of the first data stored in one or more data storage volumes in a distributed storage system is stored in the distributed cache system implemented utilizing free storage space in said distributed storage system, in response to a first computing system receiving a request to access the first data stored in the one or more data storage volumes in the distributed storage system;

instead of accessing the first data stored in the one or more data storage volumes in the distributed storage system, accessing the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored in a first data storage medium in a first cache locally associated with the first computing system

instead of accessing the first data stored in the one or more data storage volumes in the distributed storage system, requesting a second computing system, other than the first computing system, in the network to access the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored not in the first data storage medium in the first cache, but in a second data storage medium in a second cache locally associated with the second computing system; and

accessing the first data from the one or more data storage volumes in the distributed storage system, in response to determining that the copy of the first data is not stored in the distributed cache system, wherein the distributed cache system comprises portions of the first data storage medium in the first cache and portions of the second data storage medium in the second cache utilized for caching data stored in the distributed storage system,

wherein the cache metadata further comprises a mapping between the copy of the first data stored in the distributed cache system and the first data stored in the one or more data storage volumes in the distributed storage system, wherein the metadata is propagated among one or more computing systems supporting the distributed cache system to enable the first computing system to determine storage location of the first data in both the distributed cache system and the one or more data storage volumes in the distributed storage system, and to provide for the first data stored in the one or more data storage volumes in the distributed storage system to be updated when the copy of the first data stored in the distributed cache system is updated.

2. The method of claim 1, wherein the first data is accessed from the distributed storage system, further comprising determining whether the distributed cache system has sufficient storage space available to store the copy of the first data.

3. The method of claim 2, further comprising:

storing the copy of the first data in the first data storage medium at the first computing system, in response to determining that the first storage medium has sufficient memory available to store the copy of the first data; and

requesting a third computing system in the network to store the copy of the first data in a third data storage medium at the third computing system, in response to determining that the first data storage medium has insufficient storage space available to store the copy of the first data while the third data storage medium has sufficient memory available to store the copy of the first data.

4. The method of claim 3, wherein the distributed cache system further comprises metadata identifying the third computing system and respective amount of storage space available at the third computing system.

5. The method of claim 2, further comprising removing a copy of second data stored in the distributed storage system from the distributed cache system prior to storing the copy of the first data in the distributed cache system, in response to determining that the distributed cache system has insufficient memory available to store the copy of the first data.

6. The method of claim 1, wherein the distributed storage system is configured to have a primary volume and one or more secondary volumes associated with the primary volume.

7. The method of claim 6, wherein copies of data stored in the primary volume are retained in the distributed cache system until a first event occurs.

8. The method of claim 7, wherein the distributed cache system further comprises metadata for distinguishing between the copies of data stored in the primary volume and copies of data stored in the secondary volumes.

9. The method of claim 1, wherein computing systems in the network satisfy one or more security conditions.

10. The method of claim 9, wherein access to data stored in the distributed storage system is limited to the computing systems in the network.

11. The method of claim 1, further comprising implementing a cache coherency protocol to prevent simultaneous access to the shared data by different processes.

12. The method of claim 1, wherein one or more of the computing systems in the network are virtual machines such that the distributed storage system comprises virtual memories allocated to the virtual machines.

13. A computer-implemented method for accessing data stored in a distributed storage system, the method comprising:

maintaining a distributed storage system comprising a plurality of virtual machines executed on a plurality of computing systems connected over a network, wherein a portion of a respective data storage volume on each of the plurality of computing systems is allocated to one of the plurality of virtual machines executed on the respective computing system as a respective virtual memory,

wherein upon addition of each of the plurality of computing systems to the network, determining whether there is sufficient free storage space on the respective computing system to be allocated to a distributed cache system implemented over the distributed storage system, the plurality of computing systems sharing free space in the virtual memories allocated to the plurality of virtual machines based on cache metadata identifying the amount of free storage space available on one or more of said plurality of computing systems;

determining, based on metadata associated with first data, whether a copy of first data in one or more data storage volumes in a distributed storage system is stored in a distributed cache system implemented utilizing free storage space in said distributed storage system, in response to a first virtual machine (VM) receiving a request to access the first data stored in the one or more data storage volumes in the distributed storage system;

instead of accessing the first data in the one or more data storage volumes in the distributed storage system, accessing the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored in a first virtual memory associated with the first VM in the network;

instead of accessing the first data in the one or more data storage volumes in the distributed storage system, requesting a second VM in the network to access the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored not in the first virtual memory, but in a second virtual memory associated with the second VM; and

accessing the first data from the one or more data storage volumes in the distributed storage system, in response to determining that the copy of the first data is not stored in the distributed cache system wherein the distributed cache system comprises portions of the first virtual memory and portions of the second virtual memory utilized for caching data stored in the distributed storage system,

wherein the cache metadata wherein the distributed cache system further comprises a mapping between the copy of the first data stored in the distributed cache system and the first data stored in the one or more data storage volumes in the distributed storage system, wherein the metadata is propagated among one or more VMs supporting the distributed cache system to enable the first VM to determine storage location of the first data in both the distributed cache system and the one or more data storage volumes in the distributed storage system, and to provide for the first data stored in the one or more data storage volumes in the distributed storage system to be updated when the copy of the first data stored in the distributed cache system is updated.

14. The method of claim 13, wherein the first data is accessed from the distributed storage system, further comprising determining whether the distributed cache system has sufficient storage space available to store the copy of the first data.

15. The method of claim 14, further comprising:

storing the copy of the first data in the first virtual memory allocated to the first VM, in response to determining that the first virtual memory has sufficient memory available to store the copy of the first data; and

requesting a third VM in the network to store the copy of the first data in a third virtual memory allocated to the third VM, in response to determining that the first virtual memory has insufficient storage space available to store the copy of the first data while the third virtual memory has sufficient memory available to store the copy of the first data.

16. The method of claim 15, wherein the distributed cache system further comprises metadata identifying the third VM and respective amount of storage space available at the third VM.

17. The method of claim 14, further comprising removing a copy of second data stored in the distributed storage system from the distributed cache system prior to storing the copy of the first data in the distributed cache system, in response to determining that the distributed cache system has insufficient memory available to store the copy of the first data.

18. The method of claim 14, wherein the distributed storage system is configured to have a primary volume and one or more secondary volumes associated with the primary volume.

19. The method of claim 18, wherein copies of data stored in the primary volume are retained in the distributed cache system until a first event occurs.

20. The method of claim 19, wherein the distributed cache system further comprises metadata for distinguishing between the copies of data stored in the primary volume and copies of data stored in the secondary volumes.

21. The method of claim 13, wherein VMs in the network satisfy one or more security conditions.

22. The method of claim 21, wherein access to data stored in the distributed storage system is limited to the VMs in the network.

23. The method of claim 13, further comprising implementing a cache coherency protocol to prevent simultaneous access to shared data by different processes.

24. A computer program product comprising a non-transitory data storage medium having logic code stored thereon, wherein the logic code when executed on a computer causes the computer to:

maintain a distributed storage system comprising a plurality of virtual machines executed on a plurality of computing systems connected over a network, wherein a portion of a respective data storage volume on each of the plurality of computing systems is allocated to one of the plurality of virtual machines executed on the respective computing system as a respective virtual memory,

wherein upon addition of each of the plurality of computing systems to the network, it is determined whether there is sufficient free storage space on the respective computing system to be allocated to a distributed cache system implemented over the distributed storage system, the plurality of computing systems sharing free space in the virtual memories allocated to the plurality of virtual machines based on cache metadata identifying the amount of free storage space available on one or more of said plurality of computing systems;

determine, based on metadata associated with first data, whether a copy of first data in one or more data storage volumes in a distributed storage system is stored in a distributed cache system implemented utilizing free storage space in said distributed storage system, in response to a first computing system receiving a request to access the first data stored in the distributed storage system;

instead of accessing the first data in the one or more data storage volumes in the distributed storage system, access the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored in a first data storage medium designated for or locally connected to the first computing system;

instead of accessing the first data in the one or more data storage volumes in the distributed storage system, request a second computing system in the network to access the copy of the first data from the distributed cache system, in response to determining that the copy of the first data is stored not in the first data storage medium, but in a second data storage medium designated for or locally connected to the second computing system; and

access the first data from the one or more data storage volumes in the distributed storage system, in response to determining that the copy of the first data is not stored in the distributed cache system, wherein the distributed cache system comprises portions of the first data storage medium and portions of the second data storage medium utilized for caching data stored in the distributed storage system,

wherein the distributed cache system further comprises metadata providing a one-to-one mapping between the copy of the first data stored in the distributed cache system and the first data stored in the one or more data storage volumes in the distributed storage system to enable the first data stored in the distributed storage system to be updated when the copy of the first data stored in the distributed cache system is updated,

wherein the cache metadata further comprises a mapping between the copy of the first data stored in the distributed cache system and the first data stored in the one or more data storage volumes in the distributed storage system, wherein the metadata is propagated among one or more computing systems supporting the distributed cache system to enable the first computing system to determine storage location of the first data in both the distributed cache system and the one or more data storage volumes in the distributed storage system, and to provide for the first data stored in the one or more data storage volumes in the distributed storage system to be updated when the copy of the first data stored in the distributed cache system is updated.