System and method for reliably replicating data

• US 20030023618A1
• Filed: 07/26/2001
• Published: 01/30/2003
• Est. Priority Date: 07/26/2001
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Garg, Praerit, Raman, Balan Sethu, Chitre, Sudarshan A., Orbits, David A.
• Granted Patent: US 6,917,951 B2
• Time in Patent Office: Days
• US Class Current: 707/203
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 16/1787 Details of non-transparently synchronising file systems
• CPC Class Codes: G06F 16/184 implemented as replicated file system
• CPC Class Codes: G06F 2201/82 Solving problems relating to consistency ensuring consistency in mirrored systems G06F11/2064
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99954 Version management

Abstract

Described is a system and method for replicating each of a set of resources to a subject computer in a replica set prior to making use of a resource in the set of resources. The set of resources includes resources that are dependent upon each other for a proper functioning of the group. A manifest file that identifies each resource in a group of interrelated resources is used. The manifest file is generated at one computer in the replica set (typically the computer at which a modification to one of the interrelated resources occurred). When the modification occurs to one of the set of resources, the manifest file is transmitted (e.g., itself replicated) to each computer in the replica set. The manifest file includes an indicator that identifies the manifest file as a special file. When received at another computer in the replica set, a service evaluates the manifest file to identify whether the appropriate versions of the identified resources exist at the receiving computer. If not, the service at that computer awaits the receipt of each resource. The wait period may include delaying the execution of an application associated with (or even included within) the manifest file from launching. Alternatively, the FRS could simply disallow access to one or more resources identified in the manifest file until all have arrived. When all have arrived, the FRS releases control of the identified resources, which may then operate or be accessed in the ordinary manner.

Claims

1. A computer-implemented method of replicating data from a first member of a replica set to a second member of a replica set, comprising:

creating a manifest file at the first member, the manifest file including an identifier for each of a plurality of resources that exist at the first member;

causing the manifest file to be reproduced at the second member of the replica set;

in response to the manifest file being reproduced at the second member, identifying whether each resource identified in the manifest file exists at the second member; and

if a first resource identified in the manifest file does not exist at the second member, preventing a second resource identified in the manifest file from being executed until the first resource does exist at the second member.

2. The computer-implemented method of claim 1, wherein identifying whether each resource exists at the second member includes comparing information in the manifest file with information stored at the second member, the information stored at the second member identifying a plurality of resources stored at the second member.

3. The computer-implemented method of claim 1, wherein the identifier for each resource includes a version identifier associated with the resource.

4. The computer-implemented method of claim 3, wherein identifying whether each resource exists at the second member includes comparing the version identifier for the resource with another version identifier associated with another copy of the resource stored at the second member.

5. The computer-implemented method of claim 1, further comprising if the first resource does not exist at the second member, awaiting receipt of the first resource at the second member and, in response to receiving the first resource at the second member, executing the second resource.

6. The computer-implemented method of claim 1, further comprising if the first resource does not exist at the second member, awaiting receipt of every resource identified in the manifest file, and in response to a final resource identified in the manifest file being received at the second member, executing the second resource.

7. A computer-readable medium on which is stored a data structure, the data structure comprising:

a first field identifying the data structure as a special data structure; and

a second field identifying a plurality of resources, at least two of the plurality of resources being necessary for a proper functioning of the plurality of resources,

wherein, in response to determining that the data structure is a special data structure based on the information stored in the first field, a service determines whether each of the plurality of resources exists on the computer-readable medium, and if not, delays access to the plurality of resources.

8. The computer-readable medium of claim 7, wherein the data structure further comprises a third field identifying a security context in which the plurality of resources may function, the security context being similar to a security context associated with a creator of the data structure

9. The computer-readable medium of claim 8, wherein the service launches at least one of the plurality of resources identified in the second field in a process having the security context identified in the third field.

10. The computer-readable medium of claim 7, wherein the data structure is transmitted over a transmission medium from the computer-readable medium to another computer-readable medium having another instance of the service.

11. The computer-readable medium of claim 7, wherein the data structure further comprises a third data filed including a time at which the manifest file will expire.

12. The computer-readable medium of claim 11, wherein the expiration time identifies a time within which each of the plurality of resources must be available on the computer-readable medium.

13. The computer-readable medium of claim 7, wherein the data structure further comprises a third field identifying a launch-mode option for executing a particular resource within the plurality of resources.

14. The computer-readable medium of claim 13, wherein each resource in the plurality of resources includes a version identifier, and wherein the launch-mode option indicates that the particular resource should be executed if the version identifiers satisfy a selected criteria.

15. The computer-readable medium of claim 14, wherein the selected criteria indicates that the particular resource should be executed only if each version identifier associated with existing resources matches version identifiers specified in the data structure.

16. The computer-readable medium of claim 14, wherein the selected criteria indicates that the particular resource should be executed if each version identifier associated with existing resources is at least as current as version identifiers specified in the data structure.

17. The computer-readable medium of claim 14, wherein the selected criteria indicates that another resource, other than the particular resource, should be executed if a required condition is not met.

18. A computer-readable medium having computer-executable instructions for facilitating the replication of data from a first member of a replica set to a second member of the replica set, comprising:

receiving a notice that a resource in a group of resources is being modified, the group of resources being interrelated such that a proper functioning of the group of resources is dependent on a similar version of each resource in the group of resources coexisting;

in response to the notice, issuing an instruction to create a manifest file; and

adding to the manifest file an identifier for each resource in the group of resources.

19. The computer-readable medium of claim 18, wherein adding the identifier for each resource to the manifest file further comprises adding to the manifest file a globally-unique identifier for each resource.

20. The computer-readable medium of claim 18, wherein adding the identifier for each resource to the manifest file further comprises adding to the manifest file a version identifier for each resource.