Scalable distributed object management in a distributed fixed content storage system

• US 20080126404A1
• Filed: 08/28/2006
• Published: 05/29/2008
• Est. Priority Date: 08/28/2006
• Status: Active Grant

• Current Assignee: NetApp, Inc.
• Original Assignee: Bycast Inc. (NetApp, Inc.)
• Inventors: Seiler, Oliver, Lampert, Markus, Slik, David, Montour, Mike
• Granted Patent: US 7,546,486 B2
• Time in Patent Office: Days
• US Class Current: 707/103R
• CPC Class Codes: G06F 16/1774 Locking methods, e.g. locking methods for file systems allowing shared and concurrent access to files
• CPC Class Codes: G06F 16/178 Techniques for file synchronisation in file systems
• CPC Class Codes: Y10S 707/99944 Object-oriented database structure

Abstract

A method and apparatus for the management of the placement of immutable fixed-content objects in a distributed, multi-site, multi-tier non-volatile storage hierarchy consisting of a plurality of storage media connected together over networks of varying bandwidth where the computing, storage, and network resources are subject to failure. The disclosure provides for scalability, performance, and robustness by distributing the object placement management tasks amongst a plurality of computing resources and eliminating the need for global synchronization or locking amongst said computing resources.

Claims

1. A method of maintaining fixed-content objects in a storage system, comprising:

receiving a fixed-content object to be stored in accordance with information storage management policies;

storing the fixed-content object to a first storage component;

storing object management data associated with the fixed-content object to a first object management component;

identifying at the first object management component that an object management routine associated with the first management component is authorized to make changes to object management data associated with the fixed-content object;

storing the fixed-content object to at least a second storage component;

storing the object management data associated with the fixed-content object to at least a second object management component;

identifying at the second object management component that the object management routine associated with the first object management component is authorized to make changes to the object management data associated with the fixed-content object;

identifying at the second object management component that the object management routine associated with the second object management component is authorized to provide information related to the fixed-content object, but is not authorized to make changes to the object management data associated with the fixed-content object;

identifying at the first object management component that the object management routine associated with the second object management component is authorized to provide information related to the fixed-content object, but is not authorized to make changes to the object management data associated with the fixed-content object;

after detecting a failure in communication between the first object management component and the second object management component, identifying at the second object management component that an object management routine associated with the second object management component is authorized to make changes to the object management data associated with the fixed-content object; and

after detecting restoration of communication between the first object management component and the second object management component, synchronizing the first object management component and the second object management component so that only one of the object management components is authorized to make changes to the object management data associated with the fixed-content object.

2. The method of claim 1, wherein the first object management component stores object management data associated with a first plurality of fixed-content objects, and the second object management component stores object management data associated with a second plurality of fixed-content objects, wherein a subset of the first plurality of fixed-content objects corresponds to a subset of the second plurality of fixed-content objects.

3. The method of claim 1, wherein the object management data comprises object lifecycle management information.

4. The method of claim 1, wherein the object management data comprises an object location.

5. The method of claim 1, further comprising storing the fixed-content object to a plurality of additional storage components.

6. The method of claim 1, further comprising storing object management data to a plurality of additional object management components and identifying at each of the additional object management components that the object management routine associated with the first object management component is authorized to make changes to the object management data associated with the fixed-content object.

7. The method of claim 1, further comprising:

storing the object management data associated with the fixed-content object to a third object management component;

identifying at the third object management component that the object management routine associated with the second object management component is authorized to make changes to the object management data associated with the fixed-content object; and

identifying at the third object management component that the object management routine associated with the third object management component is authorized to provide information related to the fixed-content object, but is not authorized to make changes to the object management data associated with the fixed-content object.

8. The method of claim 7, further comprising:

synchronizing the third object management component so that only one of the object management components is authorized to make changes to the object management data associated with the fixed-content object.

9. The method of claim 1, further comprising:

receiving at an object management component a request to make a change to the object management data associated with the fixed-content object;

determining which object management component is authorized to make the change to the object management data associated with the fixed-content object; and

forwarding the request to the object management component that is authorized to make the change.

10. The method of claim 9, wherein detecting a failure in communication occurs when forwarding the request to the object management component that is authorized to make the change fails.

44. A computer readable medium having stored therein instructions for causing a processor to execute the method of claim 1.

11. A system that maintains fixed-content objects, comprising:

a first storage server configured to store fixed-content objects;

a second storage server configured to store fixed-content objects;

a first object management server that that stores object management data associated with a fixed-content object, the object management data comprising identification of an object management server that is authorized to make changes to the object management data;

a second object management server that that stores object management data associated with the fixed-content object, the object management data comprising identification of an object management server that is authorized to make changes to the object management data

wherein the second object management server is configured to detect the status of the first object management server, and to identify the second object management server as the object management server that is authorized to make changes to the object management data after detecting the first object management server is not accessible.

12. The system of claim 11, wherein the second object management server is configured to identify the first object management server as the object management server that is authorized to make changes to the object management data after detecting the first object management server is again accessible.

13. The system of claim 11, wherein the first object management server is configured to identify the second object management server as the object management server that is authorized to make changes to the object management data after detecting the first object management server and second object management server are in communication.

14. A method of managing synchronization of one or more attributes corresponding to a stored fixed-content object, comprising:

storing a fixed-content object in at least a first storage component and a second storage component, wherein the fixed-content object stored at the second storage component is a copy of the fixed-content object stored at the first storage component;

storing, at a first object management component, object-management-information that identifies the first object management component as a primary object management component for the fixed-content object and at least a second object management component as a secondary object management component for the fixed-content object;

storing, at the second object management component, object-management-information that identifies the first object management component as the primary object management component for the fixed-content object and the second object management component as the secondary object management component for the fixed-content object;

detecting that the first object management component is no longer accessible by the second object management component; and

modifying the object-management-information stored at the second object management component to identify the second object management component as a primary object management component for the fixed-content object.

15. The method of claim 14, further comprising:

detecting that the first object management component and the second object management component are both primary object management components.

16. The method of claim 15, further comprising:

modifying the object-management-information stored at the second object management component to identify the first object management component as the primary object management component.

17. The method of claim 15, further comprising:

synchronizing an attribute stored at the second object management component with a corresponding attribute stored at the first object management component.

18. The method of claim 15, wherein detecting that the first object management component and the second object management component are both primary object management components comprises transmitting a communication from the first object management component to the second object management component and receiving in response a communication from the second object management component to the first object management component that identifies the second object management component as a primary object management component.

19. The method of claim 14, further comprising:

determining that the first object management component is accessible again by the second object management component.

20. The method of claim 19, wherein determining that the first object management component is accessible by the second object management component comprises transmitting a communication from the second object management component to the first object management component and receiving in response a communication from the first object management component to the second object management component.

21. The method of claim 19, further comprising:

synchronizing the object-management-information stored at the first and second object management components to identify one object management component as the primary object management component.

22. The method of claim 19, further comprising:

synchronizing an attribute stored at the second object management component with a corresponding attribute stored at the first object management component.

23. The method of claim 14, further comprising

modifying an attribute corresponding to the second fixed-content object that is stored at the second storage component;

determining that the first object management component is again accessible by the second object management component, and detecting that the first object management component and the second object management component are both primary object management components; and

synchronizing the modified attribute stored at the second object management component with a corresponding attribute stored at the first object management component.

24. The method of claim 14, further comprising:

storing, at a third object management component, object-management-information that identifies the first object management component as a primary object management component for the fixed-content object and the third object management component as a secondary object management component for the fixed-content object;

determining that the first object management component is not accessible by the second object management component, but the third object management component is accessible by the second object management component; and

modifying the object-management-information stored at the third object management component to identify the second object management component as the primary object management component.

25. The method of claim 24, further comprising:

determining that the first object management component is again accessible by the second object management component, and detecting that the first object management component and the second object management component are both primary object management components for the fixed-content object; and

modifying the object-management-information stored at the second object management component and the third object management component to identify the first object management component as the primary object management component for the fixed-content object.

26. The method of claim 24, further comprising modifying an attribute corresponding to the fixed-content object that is stored at the second object management component, and replicating the modified attribute to the third object management component.

27. The method of claim 14, further comprising:

storing, at a third object management component, object-management-information that identifies the second object management component as a primary object management component for the fixed-content object and the third object management component as a secondary object management component for the fixed-content object;

determining that the first object management component is again accessible by the second object management component; and

modifying the object-management-information stored at the second and third object management components to identify the first object management component as the primary object management component.

28. The method of claim 14, wherein at least one of the attributes comprises metadata information.

29. The method of claim 14, wherein at least one of the attributes comprises an object location.

30. The method of claim 14, wherein at least one of the attributes comprises object lifecycle management information.

31. The method of claim 30, wherein the lifecycle management information is selected from the group comprising notification information, placement information, caching information, and retention time enforcement information.

32. A system of maintaining objects in a distributed storage system, comprising:

a plurality of distributed storage nodes that store objects;

a plurality of distributed storage management nodes in communication with the storage nodes over a network, the storage management nodes configured to maintain management information about the stored objects;

wherein a first storage management node is configured to fulfill requests to change storage management information about an object and other storage management nodes are configured to forward requests to change storage management information about the object to the first storage management node.

33. The system of claim 32, further wherein the first storage management node is configured to communicate the changed storage management information to other storage management nodes.

34. The system of claim 32, wherein a second storage management node is configured to fulfill requests to change storage management information about the object upon detecting that the first storage management node is unreachable.

35. The system of claim 34, wherein the second storage management node is configured to direct future requests to change storage management information about the object from other reachable storage management nodes to the second storage management node.

36. The system of claim 34, wherein the first storage management node is configured to synchronize with the second storage management node upon detecting that the second storage management node is configured to receive requests from other storage management nodes.

37. A method of maintaining fixed-content objects in a storage system, comprising:

receiving a fixed-content object to be stored in accordance with information storage management policies;

storing the fixed-content object to a plurality of storage components;

storing object management data associated with the fixed-content object to a plurality of object management components;

identifying at the plurality of object management components that the object management routine associated with any of the plurality of object management component is authorized to provide information related to the fixed-content object, but only an object management routine associated with a first object management component is authorized to make changes to the object management data associated with the fixed-content object; and

after detecting the first object management component is no longer accessible, identifying at one or more of the remaining plurality of object management components that an object management routine associated with a second object management component is authorized to make changes to the object management data associated with the fixed-content object.

38. The method of claim 37, further comprising detecting the first object management component is again accessible and synchronizing the first object management component and the second object management component so that only one of the object management components is authorized to make changes to the object management data associated with the fixed-content object.

39. The method of claim 37, further comprising detecting the second object management component is no longer accessible and identifying at one or more of the remaining plurality of object management components that an object management routine associated with a third object management component is authorized to make changes to the object management data associated with the fixed-content object.

40. The method of claim 37, wherein the first object management component is no longer accessible because of a network communication failure.

41. The method of claim 37, wherein the first object management component is no longer accessible because the first object management component is offline.

42. The method of claim 37, wherein the first object management component is no longer accessible because the first object management component was removed from the storage system.

43. A method of maintaining fixed-content objects in a storage system, comprising:

receiving a fixed-content object to be stored in accordance with information storage management policies;

storing the fixed-content object to a plurality of storage components;

storing object management data associated with the fixed-content object to a plurality of object management components;

identifying at the plurality of object management components that the object management routine associated with any of the plurality of object management component is authorized to provide information related to the fixed-content object, but only an object management routine associated with a first object management component is authorized to make changes to the object management data associated with the fixed-content object; and

identifying at the plurality of object management components that only an object management routine associated with a second object management component is authorized to make changes to the object management data associated with the fixed-content object after determining that it is not optimal to use the first object management component to make changes to the object management data.