METADATA SERVER AND METADATA MANAGEMENT METHOD

• US 20100161657A1
• Filed: 07/16/2009
• Published: 06/24/2010
• Est. Priority Date: 12/18/2008
• Status: Abandoned Application

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: CHA, Myung Hoon, KIM, Hong Yeon, JIN, Ki Sung, KIM, Young Kyun, NAMGOONG, Han
• Application Number:
  US12/504,621
• Publication Number:
  US 20100161657A1
• Time in Patent Office: Days
• US Class Current: 707/770
• CPC Class Codes: G06F 11/1662 the resynchronized component or unit being a persistent storage device re-synchronization of failed mirror storage G06F11/2082; rebuild or reconstruction of parity RAID storage G06F11/1008
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• 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 16/134 Distributed indices

Abstract

Provided are a metadata server cluster and a metadata management method thereof, which distribute metadata for a file to a cluster including a plurality of metadata servers to replicate the metadata. The metadata server includes a directory hierarchy storage unit, a metadata storage unit, and a search unit. The directory hierarchy storage unit stores all directory hierarchies which are stored in the metadata server cluster. The metadata storage unit stores metadata for a data file. The search unit searches the directory hierarchies and the metadata.

Claims

1. A metadata server in a metadata server cluster including a plurality of metadata servers sharing a directory hierarchy, the metadata server comprising:

a directory hierarchy storage unit storing all directory hierarchies which are stored in the metadata server cluster;

a metadata storage unit storing metadata for a data file; and

a search unit searching the directory hierarchies and the metadata.

2. The metadata server of claim 1, further comprising a directory hierarchy duplication unit duplicating the directory hierarchy to other metadata server of the metadata server cluster.

3. The metadata server of claim 1, further comprising:

a metadata duplication unit duplicating a content of the metadata storage unit to at least one other metadata server of the metadata server cluster; and

a recovery unit detecting a failure of the metadata server to recover the detected failure.

4. The metadata server of claim 3, wherein when a failure of a metadata server where metadata are duplicated is detected, the recovery unit duplicates the metadata in a new metadata server.

5. The metadata server of claim 1, wherein the metadata storage unit stores metadata for a plurality of data files of some directories among a plurality of directories which are stored in the metadata server cluster.

6. The metadata server of claim 1, wherein the search unit checks a metadata server storing metadata for a queried data file through the directory hierarchy storage unit in response to a data file inquiry query, and searches a requested metadata through the metadata storage unit in response to a metadata request.

7. A metadata management method in a metadata server cluster including a plurality of metadata servers, the metadata management method comprising:

checking a metadata server storing metadata for a data file on which a client requests an inquiry;

providing information of the checked metadata server to the client;

searching, by the checked metadata server, the metadata for the requested data file in response to the inquiry request of the client; and

providing the searched metadata to the client.

8. The metadata management method of claim 7, wherein the checking of a metadata server comprises:

receiving, by an arbitrary metadata server in the metadata server cluster, a data file inquiry request from the client;

searching, by a metadata server receiving the inquiry request, a directory hierarchy stored in the metadata server to check a directory comprising the data file; and

checking a metadata server storing metadata for a plurality of files of the checked directory.

9. The metadata management method of claim 8, wherein the arbitrary metadata server is selected at random by the client.

10. The metadata management method of claim 7, wherein the searching of the metadata comprises:

checking, by the checked metadata server, a directory comprising a data file inquired by the client among a plurality of directories which are stored in the checked metadata server; and

searching metadata for the data file in the checked directory.

11. The metadata management method of claim 7, further comprising duplicating the same metadata in other metadata server of the metadata server cluster.

12. The metadata management method of claim 11, wherein the duplicating of the same metadata comprises:

selecting metadata to duplicate;

selecting one or more metadata servers to store a copy of the metadata;

duplicating the selected metadata in the one or more metadata servers; and

reflecting the duplicated content in the directory hierarchy.

13. The metadata management method of claim 7, further comprising detecting a failure of the metadata server to recover the detected failure.

14. The metadata management method of claim 13, wherein the recovering of the detected failure comprises:

recognizing a metadata server where a failure occurs, among a plurality of metadata servers;

checking metadata which are stored in the metadata server having the failure;

recognizing a specific metadata server storing the same copy as the checked metadata; and

duplicating the copy in other metadata server.

15. The metadata management method of claim 7, further comprising adding a new metadata server in the metadata server cluster.

16. The metadata management method of claim 15, wherein the adding of a new metadata server comprises:

registering information of the new metadata server in the metadata server cluster; and

duplicating a directory hierarchy, which is stored in an existing metadata server of the metadata server cluster, in the new metadata server.

17. A metadata management method in a metadata server cluster including a plurality of metadata servers, the metadata management method comprising:

receiving a making request of a directory;

selecting a metadata sever to make the directory;

making a directory in a metadata storage unit of the selected metadata server; and

reflecting the directory making to update a directory hierarchy.

18. The metadata management method of claim 17, wherein the updating of a directory hierarchy comprises:

updating, by one metadata server comprised in the metadata server cluster, a directory hierarchy of the metadata server; and

duplicating the updated directory hierarchy to other metadata servers of the metadata server cluster.

19. A metadata management method in a metadata server cluster including a plurality of metadata servers, the metadata management method comprising:

receiving a metadata file generation request from a client;

checking a metadata server with a directory to which the data file will belong;

notifying the client of information on the checked metadata server; and

receiving, by the checked metadata server, the metadata file generation request from the client, and generating a metadata file in the metadata server in response to the generation request.

20. The metadata management method of claim 19, wherein the checking of a metadata server is performed by an arbitrary metadata server of the metadata server cluster which is selected at random by the client.