Distributed, unified file system operations

• US 9,037,618 B2
• Filed: 03/31/2011
• Issued: 05/19/2015
• Est. Priority Date: 03/31/2011
• Status: Expired due to Fees

• Current Assignee: Micro Focus Software, Inc. (Open Text Corporation)
• Original Assignee: Novell Incorporated (Open Text Corporation)
• Inventors: Srinivasaraghavan, Haripriya, Madhavan, Vijai Babu
• Primary Examiner(s): Jacob, Ajith
• Application Number:
  US13/077,248
• Publication Number:
  US 20120254140A1
• Time in Patent Office: 1,510 Days
• Field of Search: 707/822
• US Class Current: 707/822
• CPC Class Codes: G06F 11/1435 using file system or storage system metadata
• CPC Class Codes: G06F 16/1844 Management specifically adapted to replicated file systems
• CPC Class Codes: G06F 16/9014 hash tables

Abstract

Apparatus, systems, and methods may operate to receive a request to access a file from a client node at one of a plurality of lookup nodes, the lookup nodes storing portions of a distributed table having lookup information associated with the file. The distributed table, which may comprise a distributed hash table, includes replicated references to virtualized and non-virtualized file elements of varied granularity in a unified file system. Additional activity may comprise forwarding the request to another one of the lookup nodes until the lookup information is located, or chaining the request to one of the plurality of lookup nodes and implementing activity that has been requested. Additional apparatus, systems, and methods are disclosed.

Claims

1. A system, comprising:

a data server to perform operations on a file of a unified file system;

a distributed table having lookup information for the file, the distributed table supporting the unified file system, the lookup information being separate from the unified file system, the lookup information describing the location of the data server;

a plurality of lookup nodes to store portions of the distributed table, no single portion being an entirety of the distributed table, and the portions being are distributed evenly among the plurality of lookup nodes,

a lookup node in the plurality of lookup nodes, the lookup node comprising;

a memory to store policies that direct a lookup path through the plurality of lookup nodes; and

a processor to;

receive a request from a client node to access the file, the request including the file name and a path of the file;

generate a key for the file by hashing the file name and the path of the file received in the request;

determine that the lookup information for the file is not stored on the lookup node by looking up the key the portion of the distributed table resident on the lookup node; and

forward the request to another one of the plurality of lookup nodes according to the policies stored in the memory to chain the request to the plurality of lookup nodes to find a second lookup node, from the plurality of lookup nodes, with the lookup information so that the access that has been requested can be directly implemented by the data server.

2. The apparatus of claim 1, further comprising:

a user input device operating as the client node.

3. The apparatus of claim 1, wherein the lookup nodes comprise a combination of file servers and file meta-data servers.

4. The apparatus of claim 1, wherein the client node is housed together with at least one of the lookup nodes.

5. A processor-implemented method to execute on one or more processors that perform the method, comprising:

receiving, at a lookup node from a plurality of lookup nodes, a request to access a file of a unified file system from a client node, the request including a file name and a path of the file, the lookup nodes storing portions of a distributed table, the distributed table having lookup information for the file, wherein the distributed table includes replicated references to virtualized and non-virtualized file elements of varied granularity in a unified file system, no single portion of the distributed table is an entirety of the distributed table, the portions of the distributed table being distributed evenly among the plurality of lookup nodes, the lookup information being separate from the unified file system and describing the location of a data server that can perform operations on the file;

generate, by the lookup node, a key for the file by hashing the file name and the path of the file;

determining, by the lookup node, that the lookup information for the file is not stored on the lookup node by looking up the key in a portion of the distributed table resident on the lookup node;

accessing, by the lookup node, a policy to determine a second lookup node from the plurality of lookup nodes to which the request will be forwarded; and

forwarding, by the lookup node, the request to the second lookup node to chain the request to the plurality of lookup nodes to find the lookup information and implement, by the data server, the access that has been requested.

6. The method of claim 5, further comprising:

transmitting the lookup information to the client node, wherein the lookup information permits the client node to access the file via the data server.

7. The method of claim 5, wherein a reference in the table comprises a location of the lookup information, or a location of a subsequent one of the lookup nodes.

8. The method of claim 5, further comprising:

redirecting the client node to a second data server when the data server, referenced by the lookup information, determines that the lookup information refers to one of the virtualized file elements.

9. The method of claim 5, wherein at least one of the replicated references to one of the virtualized file elements comprises one of a symbolic link or an extended attribute.

10. The method of claim 5, wherein a first subset of the replicated references are formatted according to a first protocol, and wherein a second subset of the replicated references are formatted according to a second protocol different from the first protocol.

11. The method of claim 5, wherein the unified file system comprises a unified single logical namespace to virtualize distributed volumes and/or individual file locations.

12. The method of claim 5, further comprising:

partitioning the table among the plurality of lookup nodes so that some entries in the table in each one of the lookup nodes overlap entries in the table in at least one of the other lookup nodes.

13. The method of claim 5, further comprising:

adding an additional lookup node to the plurality of lookup nodes by selecting a portion of the table included in the additional lookup node and distributing the portion to at least one other one of the lookup nodes.

14. At least one non-transitory machine readable medium including instructions that, when executed by a machine, cause the machine to perform operations comprising:

receiving, at a lookup node from a plurality of lookup nodes, a request to access a file of a unified file system from a client node, the request including a file name and a path of the file, the lookup nodes storing portions of a distributed table, the distributed table having lookup information for the file, wherein the distributed table includes replicated references to virtualized and non-virtualized file elements of varied granularity in a unified file system, no single portion of the distributed table is an entirety of the distributed table, the portions of the distributed table being distributed evenly among the plurality of lookup nodes, the lookup information being separate from the unified file system and describing the location of a data server that can perform operations on the file;

generate, by the lookup node, a key for the file by hashing the file name and the path of the file;

determining, by the lookup node, that the lookup information for the file is not stored on the lookup node by looking up the key in a portion of the distributed table resident on the lookup node;

accessing, by the lookup node, a policy to determine a second lookup node from the plurality of lookup nodes to which the request will be forwarded; and

forwarding, by the lookup node, the request to the second lookup node to chain the request to the plurality of lookup nodes to find the lookup information and implement, by the data server, the access that has been requested.

15. The medium of claim 14, wherein the operations further comprise:

transmitting the lookup information to the client node, wherein the lookup information permits the client node to access the file via the data server.

16. The medium of claim 14, wherein a reference in the table comprises a location of the lookup information, or a location of a subsequent one of the lookup nodes.

17. The medium of claim 14, wherein the operations further comprise:

redirecting the client node to a second data server when the data server, referenced by the lookup information, determines that the lookup information refers to one of the virtualized file elements.

18. The medium of claim 14, wherein a first subset of the replicated references are formatted according to a first protocol, and wherein a second subset of the replicated references are formatted according to a second protocol different from the first protocol.

19. The medium of claim 14, wherein the operations further comprise:

partitioning the table among the plurality of lookup nodes so that some entries in the table in each one of the lookup nodes overlap entries in the table in at least one of the other lookup nodes.