Method and system for managing file metadata transparent about address changes of data servers and movements of their disks

• US 8,046,345 B2
• Filed: 07/22/2008
• Issued: 10/25/2011
• Est. Priority Date: 12/14/2007
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Kim, Hong-Yeon, Kim, Young-Kyun, Lee, Sang-Min, Kim, Myung-Joon
• Primary Examiner(s): Breene; John E
• Assistant Examiner(s): Ly; Anh
• Application Number:
  US12/177,635
• Publication Number:
  US 20090157694A1
• Time in Patent Office: 1,190 Days
• Field of Search: 707/706, 707/716, 707/770, 709/203, 711/114, 711/170, 711/203, 714/11
• US Class Current: 707/706
• CPC Class Codes: G06F 16/14 Details of searching files based on file metadata

Abstract

Provided are a method and a system for managing a file metadata, which are independent of a address change of a data server and movements of a disk by using a location of disk where data exists actually as a location information of a file data. The method for managing the file metadata includes determining whether the request of the metadata is valid, verifying information of a disk storing a file data corresponding to the metadata and a data server address in which the file data exists, and transmitting the verified address of the data server to the client. According to the method and the system for managing the file metadata, data services can be continuously provided without modifications of a file metadata even if the data server address is changed.

Claims

1. A method for managing file metadata, comprising:

requesting, at a client, metadata from a metadata management server;

verifying information of a disk storing file data corresponding to the metadata, and a data server address in which the file data exists when the request is valid; and

transmitting the verified address of the data server to the client,

wherein the metadata management server comprises a disk identifier directory comprising one or more disk identifier files each comprising a disk identifier, general disk information, and the data server address where the disk corresponding to the disk identifier exists.

2. The method of claim 1, wherein the general disk information comprises at least one of disk state, total disk space and used disk space.

3. The method of claim 1, wherein the verifying of the data server address comprises:

searching the disk identifier file in a disk management directory of the metadata management server by using the disk identifier of the metadata; and

extracting the data server address of the disk identifier file.

4. A metadata management server, comprising:

a metadata managing means for transmitting metadata corresponding to a client'"'"'s request to the client;

a disk identifier creation means for creating a unique disk identifier corresponding to each disk in a data server; and

a disk identifier managing means for creating and updating a disk identifier file comprising a data server address corresponding each disk identifier,

wherein the disk identifier managing means receives each disk identifier, general disk information, and a data server address from the data server, checks if the received information is identical to information of a retained disk identifier file, and updates the information of the disk identifier file with the received information when the received information is different from the information of the disk identifier file.

5. A data server comprising:

a disk identifier issuing means for determining whether a disk identifier of a disk in the data server exists and performing an issuing of the disk identifier;

a disk information managing means for periodically verifying all disk information in the data server together with a metadata management server; and

a file data managing means for transmitting a corresponding file data at a request of a client,

wherein the disk identifier managing means verifies whether the disk identifier in each registered disk list exists at the start of data service, requests the metadata management server to allocate the disk identifier when the disk identifier is absent in any disk, receives the disk identifier allocated by the request, and stores the disk identifier at a corresponding disk.