Arrangement for a distributed file system having data objects mapped independent of any data object attribute

• US 8,229,985 B2
• Filed: 02/07/2005
• Issued: 07/24/2012
• Est. Priority Date: 02/07/2005
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Turner, Bryan C., Toebes, John, Walker, Douglas Jay
• Primary Examiner(s): AL HASHEMI, SANA A
• Application Number:
  US11/051,333
• Publication Number:
  US 20060179037A1
• Time in Patent Office: 2,724 Days
• Field of Search: 707/791, 707/793, 707/803, 707/825, 707/959, 707/960
• US Class Current: 707/959
• CPC Class Codes: G06F 16/10 File systems; File servers
• CPC Class Codes: G06F 16/1844 Management specifically adapted to replicated file systems
• CPC Class Codes: Y10S 707/959 Network
• CPC Class Codes: Y10S 707/96 Object-relational

Abstract

Each data file of a distributed file system is identified by a corresponding globally-unique object identifier that is independent of any attribute of the data file. A node in the distributed file system has a file system that includes a plurality of directory entries, where each directory entry has a file name field, an attributes field, and an object identifier field configured for specifying a globally-unique object identifier. The globally-unique object identifier is universally reachable by any node of the distributed file system and uniquely identifies the data file, enabling the data file to be universally identifiable by any node based on the corresponding object identifier. The data file can be stored independently of a device file system based on providing a resolution between the data file name and its object identifier, followed by a resolution between the object identifier and a selected location for the data file.

Claims

1. A method in a first network service node of a distributed file system, the method comprising:

receiving by the first network service node, from a second network service node in the distributed file system, a request for a unique object identifier assigned to a data file that does not include directory information, the data file further not including any fully qualified path to any data file stored in the distributed file system, the unique object identifier unique within the network-based distributed file system, the distributed file system implemented as a network-based distributed file system and configured for storing copies of the data file within a plurality of storage nodes within the distributed file system at respective storage locations, the data file being reachable from any storage node that stores the data file based on the unique object identifier; and

outputting to the second network service node, by the first network service node, a response to the request that specifies the unique object identifier, the unique object identifier being an arbitrary value that is independent of any attribute of the data file and independent of any directory structure in any of the storage nodes storing the data file, wherein at least one of the storage locations storing the corresponding copy of the data file is identifiable by the second network service node based on the unique object identifier, and wherein each data file stored in the distributed file system has a corresponding unique object identifier to access the corresponding data file without use of any fully qualified path in the distributed file system.

5. The method of claim 1, wherein:

the request specifies that the data file is a new data file;

the outputting includes selecting a new identifier as the unique object identifier, and storing in a table an entry that specifies the unique object identifier for the new data file.

9. The method of claim 5, further comprising:

the first network service node storing in a second table at least one entry specifying the unique object identifier and a corresponding location as the storage location for a corresponding copy of the data file;

the first network service node receiving a query from the second network service node specifying the unique object identifier and requesting the storage location for the unique object identifier; and

the first network service node selectively supplying to the second network service node at least one of the locations specified in the at least one entry based on the unique object identifier and determined attributes.

2. A distributed file system comprising:

storage devices distributed throughout the distributed file system, each configured for storing data files at respective storage locations, the distributed file system implemented as a network-based distributed file system for storage of copies of the data files among the storage devices at their respective storage locations, wherein the data files do not include directory information, the data files further not including any fully qualified path to any of the data files;

a first network service node configured for providing resolutions to queries for unique object identifiers assigned to the data files, respectively, each unique object identifier assigned to the corresponding data file being an arbitrary value that is unique within the distributed file system, independent of any storage address on any of the storage devices, and independent of any directory structure in any of the storage devices storing a corresponding copy of the corresponding data file, each data file being reachable from any storage device that stores the corresponding copy of the data file based on the corresponding unique object identifier of the data file and without use of any fully qualified path in the distributed file system; and

a second network service node configured for determining the unique object identifier for a corresponding one of the data files based on outputting the corresponding query into the distributed file system and receiving the corresponding resolution, the second network service node further configured for obtaining a second resolution, that specifies the corresponding storage location of the corresponding copy of the one data file at one of the storage devices storing the one data file, based on the corresponding unique object identifier.

6. The file system of claim 2, wherein the second network service node is configured for supplying in the query a name of the one data file, and the first network service node is configured for providing the corresponding resolution that specifies the name of the one data file and the corresponding unique object identifier.

10. The file system of claim 6, wherein the second network service node is configured for obtaining the second resolution based on sending a second query that specifies the unique object identifier for the corresponding one data file, the first network service node configured for providing the second resolution, specifying the corresponding unique object identifier and the corresponding storage location of the one data file, based on the corresponding unique object identifier and determined attributes relevant to the second network service node.

3. A network service node of a distributed file system, the network service node comprising:

a network interface configured for receiving, from a second network service node in the distributed file system, a request for a unique object identifier assigned to a data file that does not include directory information, the data file further not including any fully qualified path to any of the data files, the unique object identifier unique within the distributed file system, the distributed file system implemented as a network-based distributed file system and configured for storing copies of the data file within a plurality of storage nodes within the distributed file system at respective storage locations, the data file being reachable from any storage node that stores the data file based on the unique object identifier; and

a resolution resource configured for outputting to the second network service node a response to the request that specifies the unique object identifier, the unique object identifier being an arbitrary value that is independent of any attribute of the data file and independent of any directory structure in any of the storage nodes storing the data file, wherein at least one of the storage locations storing the corresponding copy of the data file is identifiable by the second network service node based on the unique object identifier, and wherein each data file stored in the distributed file system has a corresponding unique object identifier to access the corresponding data file without use of any fully qualified path in the distributed file system.

7. The node of claim 3, further comprising a table configured for storing an entry, wherein:

the request specifies that the data file is a new data file; and

the resolution resource is configured for selecting a new identifier as the unique object identifier, and storing in the table said entry specifying the unique object identifier for the new data file.

11. The node of claim 7, further comprising a second table, wherein:

the resolution resource is configured for storing in the second table at least one entry specifying the unique object identifier and a corresponding location as the storage location for a corresponding copy of the data file;

in response to the network interface receiving a query from the second network service node specifying the unique object identifier and requesting the storage location for the unique object identifier, the resolution resource is configured for selectively supplying to the second network service node at least one of the locations specified in the at least one entry based on the unique object identifier and determined attributes.

4. A network service node of a distributed file system, the network service node comprising:

means for receiving, from a second network service node in the distributed file system, a request for a unique object identifier assigned to a data file that does not include directory information, the data file further not including any fully qualified path to any data file stored in the distributed file system, the unique object identifier unique within the distributed file system, the distributed file system implemented as a network-based distributed file system and configured for storing copies of the data file within a plurality of storage nodes within the distributed file system at respective storage locations, the data file being reachable from any storage node that stores the data file based on the unique object identifier; and

means for outputting to the second network service node a response to the request that specifies the unique object identifier, the unique object identifier being an arbitrary value that is independent of any attribute of the data file and independent of any directory structure in any of the storage nodes storing the data file, wherein at least one of the storage locations storing the corresponding copy of the data file is identifiable by the second network service node based on the unique object identifier, and wherein each data file stored in the distributed file system has a corresponding unique object identifier to access the corresponding data file without use of any fully qualified path in the distributed file system.

8. The node of claim 4, wherein:

the request specifies that the data file is a new data file;

the means for outputting configured for selecting a new identifier as the unique object identifier, and storing in a table an entry that specifies the unique object identifier for the new data file.

12. The node of claim 8, wherein the means for outputting is configured for:

storing in a second table at least one entry specifying the unique object identifier and a corresponding location as the storage location for a corresponding copy of the data file;

in response to the means for receiving having received a query from the second network service node specifying the unique object identifier and requesting the storage location for the unique object identifier, the means for outputting selectively supplying to the second network service node at least one of the locations specified in the at least one entry based on the unique object identifier and determined attributes.