Dynamic replication of files in a network storage system

• US 7,203,731 B1
• Filed: 12/29/2000
• Issued: 04/10/2007
• Est. Priority Date: 03/03/2000
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Coates, Joshua L., Bozeman, Patrick E.
• Primary Examiner(s): Ellis; Kevin L.
• Application Number:
  US09/753,220
• Time in Patent Office: 2,293 Days
• Field of Search: 709/213, 709/214, 709/217, 709/218, 709/219
• US Class Current: 709/214
• CPC Class Codes: G06F 16/184 implemented as replicated file system
• CPC Class Codes: G06F 16/188 Virtual file systems
• CPC Class Codes: H04L 63/0823 using certificates cryptographic mechanisms or cryptographic arrangements for entity authentication involving certificates H04L9/3263
• CPC Class Codes: H04L 63/12 Applying verification of the received information cryptographic mechanisms or cryptographic arrangements for data integrity or data verification H04L9/32
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A network storage system includes a virtual file system (“VFS”) that manages the files of the network storage system, and a storage center that stores the files. The VFS and the storage center are separated, such that a client accesses the VFS to conduct file system operations and the client accesses the storage center to upload/download files. The client accesses the network storage system through one or more storage ports. The storage center includes a plurality of distributed object storage managers (DOSMs) and a storage cluster that includes a plurality of intelligent storage nodes. The network storage system includes additional storage centers at geographically disparate locations. The network storage system uses a multi-cast protocol to maintain file information at the DOSMs regarding files stored in the intelligent storage nodes, including files stored in disparate storage centers.

Claims

1. A method comprising:

in response to one request to retrieve an object, selecting one control node to retrieve the object;

retrieving, by the one control node, the object from a storage node;

storing the object in a cache at the one control node;

in response to a subsequent request, upon determining that the subsequent request is to retrieve the same object as in a prior request, selecting a different control node; and

storing the object in another cache at the different control node, a number of control nodes each storing a respective copy of the object dependent on the number of requests to retrieve the object.

2. The method of claim 1, wherein:

the selecting of the one control node and the selecting of the different control node are accomplished by a level four switch.

3. The method of claim 1, wherein:

the selecting of the one control node and the selecting of the different control node are accomplished by a load balancing fabric.

4. The method of claim 1, wherein:

the one control node and the different control node comprise distributed object storage managers.

5. The method of claim 1, wherein:

the selecting of the one control node and the selecting of the different control node are accomplished by a switch that is capable of prioritizing transmission control protocol and user datagram protocol traffic.

6. The method of claim 1, wherein:

the selecting of the one control node and the selecting of the different control node are based on availability of the one control node and the different control node, respectively.

7. A method comprising:

in response to a plurality of requests for retrieval of an object, selecting different control nodes to retrieve the object from a storage node, the selecting being based upon determining that a subsequent request is to retrieve the same object as in a prior request; and

storing the object in respective caches in the control nodes, a number of control nodes each storing a respective copy of the object dependent on the plurality of requests for retrieval of the object.

8. The method of claim 7, wherein:

the selecting is accomplished by a level four switch.

9. A system comprising:

a load balancer to, in response to one request to retrieve an object, select one control node to retrieve the object, the one control node being capable of retrieving the object from a storage node and also being capable of storing the object in a cache at the one control node; and

the load balancer being capable of, in response to a subsequent request, upon determining that it is to retrieve the same object as in a prior request, selecting a different control node to retrieve the object, the different control node being capable of retrieving the object from the storage node and of storing the object in a cache at the different control node, a number of control nodes each storing a respective copy of the object dependent on the number of requests to retrieve the object.

10. The system of claim 9, wherein:

the load balancer comprises a level four switch.

11. The system of claim 9, wherein:

the load balancer comprises a load balancing fabric.

12. The system of claim 9, wherein:

the one control node and the different control node comprise distributed object storage managers.

13. The system of claim 9, wherein:

the load balancer comprises a switch that is capable of prioritizing transmission control protocol and user datagram protocol traffic.

14. The system of claim 9, wherein:

the load balancer is capable of selecting the one control node and the different control node based on availability of the one control node and the different control node, respectively.

15. A system comprising:

a load balancer to, in response to a plurality of requests for retrieval of an object, select different control nodes to retrieve the object from a storage node, the selecting being based upon determining that a subsequent request is to retrieve the same object as in a prior request; and

each of the control nodes being capable of storing the object in respective caches in the respective control nodes, a number of control nodes each storing a respective copy of the same object dependent on the plurality of requests for retrieval of the object.

16. The system of claim 15, wherein:

the load balancer comprises a level four switch.

17. A method comprising:

distributing a plurality of requests to retrieve an object amongst a plurality of control nodes by selecting for each of the plurality of the requests a different one of the plurality of control nodes to retrieve the object upon determining that the request is to retrieve the same object as in a prior request;

retrieving, by a control node, the object from a storage node; and

storing the object in a cache at the control node, a number of control nodes each storing a respective copy of the same object dependent on the number of requests to retrieve the object.

18. The method of claim 17, wherein:

the selecting of the one of the plurality of control nodes is accomplished by a load balancing fabric.

19. The method of claim 17, wherein:

the control nodes comprise distributed object storage managers.