File sharing system and file sharing method

• US 8,838,721 B2
• Filed: 04/15/2011
• Issued: 09/16/2014
• Est. Priority Date: 04/15/2011
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Nemoto, Jun, Ishii, Yohsuke, Sutoh, Atsushi
• Primary Examiner(s): Nguyen, Thai
• Assistant Examiner(s): BOOK, PHYLLIS A
• Application Number:
  US13/126,066
• Publication Number:
  US 20120265836A1
• Time in Patent Office: 1,250 Days
• Field of Search: 709/201, 709/203, 709/208, 709/211, 709/216, 709/219, 709/245, 709/246, 707/609, 707/610, 707/623, 707/705
• US Class Current: 709/208
• CPC Class Codes: G06F 16/14 Details of searching files based on file metadata
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 16/1744 using compression, e.g. sparse files
• CPC Class Codes: G06F 16/1767 Concurrency control, e.g. optimistic or pessimistic approaches
• CPC Class Codes: G06F 16/178 Techniques for file synchronisation in file systems
• CPC Class Codes: G06F 16/487 using geographical or spatial information, e.g. location

Abstract

In a computer system according to an embodiment of the present invention, a first slave server includes copy management information associating a master file system identifier with an identifier of a slave server including a copy of the master file system identified by the master file system identifier. The first slave server acquires a copy of a part of a master file system. The copy includes a stub-state object (file or directory). The first slave server searches the copy management information for a slave server to be queried regarding information for materializing the stub-state object. If the slave server to be queried includes the information for materialization, the first slave server materializes the stub-state object using the information for materialization acquired from the queried slave server.

Claims

1. A file sharing system comprising:

a master server configured to provide one or more master file systems each including a directory and a file each being an object; and

slave servers each configured to acquire a copy file system of a part or whole of a master file system provided by the master server, wherein;

a first slave server of the slave servers includes copy management information including entries each associating a master file system identifier, an identifier of a version of a master file system identified by the master file system identifier and an identifier of a slave server which checked out of a copy file system of a part or whole of the master file system identified by the master file system identifier;

the first slave server acquires a first copy file system of a part of a first master file system in the one or more master file systems, the first copy file system including a stub-state object;

the first slave server, using an identifier of the first master file system and an identifier of a version of the first master file system, searches the copy management information for a slave server to be queried regarding cached information for materializing the stub-state object and performs the searches simultaneously to verify the presence or absence of the cached information and issues an object acquisition to the slave server being the first to return a cache present designation, and the number of slave servers to be queried being based on a version of the copy management information;

on condition that the search result indicates a slave server to be queried of the slave servers, the first slave server queries the slave server to be queried regarding the cached information for materialization;

on condition that the slave server to be queried includes the information for materialization, the first slave server materializes the stub-state object using the information for materialization acquired from the slave server to be queried; and

the first slave server synchronizes the copy management information thereon and other copy management information on one of the master server and the slave servers in a checkout process of the first copy file system;

wherein, on condition that the information for materialization is not acquired from any slave server in the copy management information, the first slave server acquires the information for materialization from the master file server of the first master file system.

2. The file sharing system according to claim 1, wherein an entry of the copy management information comprises the master file system identifier, the slave server identifier and an identifier to identify a copy file system;

and

a single entry is assigned to one copy file system in the copy management information.

3. The file sharing system according to claim 2, wherein the first copy file system of the first slave server is a working copy of the first master file system;

and

the first slave server copies an object updated in the working copy to the first master server.

4. The file sharing system according to claim 3, wherein

the slave server to be queried is associated with a master file system of which version is the same as the version of the first master file system in the copy management information.

5. The file sharing system according to claim 4, wherein the slave server to be queried returns a response indicating that the slave server to be queried does not include the information for materialization on condition that the queried object has been updated.

6. The file sharing system according to claim 5, wherein the master server includes master copy management information;

each of the one or more slave servers includes copy management information as a copy of the master copy management information; and

each of the one or more slave servers adds an entry to the master copy management information for an acquired copy file system of a part or whole of a master file system.

7. The file sharing system according to claim 6, wherein the first slave server includes a user-settable search policy defining range of the searching;

and

the first slave server searches the copy management information for the slayer server to be queried within the search range defined by the search policy.

8. The file sharing system according to claim 7, wherein the copy management information further includes information indicating a location of a slave server;

and

the first slave server refers to the location information to determine the slave server to be queried.

9. The file sharing system according to claim 8, wherein the copy management information further includes performance information associated with the copy file system identifier and indicating access performance to a copy file system identified by the copy file system identifier;

and

the first slave server refers to the performance information to determine the slave server to be queried.

10. The file sharing system according to claim 9, wherein the first slave server makes the query to the first slave server on condition that the search result indicates the first slave server as the slave server to be queried.

11. A file sharing method, in a computer system comprising a master server configured to provide one or more master file systems each including a directory and a file each being an object, and slave servers each configured to acquire a copy file system of a part or whole of a master file system provided by the master server, for a first slave server to acquire a copy file system of a part or whole of a master file system in the one or more master file systems, the file sharing method comprising:

acquiring, by the first slave server, a first copy file system of a part of a first master file system, the first copy file system including a stub-state object,

referring to, by the first slave server, copy management information including entries each associating a master file system identifier, an identifier of a version of a master file system identified by the master file system identifier and an identifier of a slave server which checked out of a copy file system of a part or whole of the master file system identified by the master file system identifier,

using, by the first slave server, an identifier of the first master file system and an identifier of a version of the first master file system, and searching the copy management information for a slave server to be queried regarding cached information for materializing the stub-state object, and performing the searches simultaneously to verify the presence or absence of the cached information and issuing an object acquisition to the slave server being the first to return a cache present designation, and the number of slave servers to be queried being based on a version of the copy management information;

receiving, by the first slave server, a query regarding information for materializing the stub-state object; and

returning, by the first slave server, a response indicating that the first slave server does not include the information for materialization on condition that the queried object has been updated;

on condition that the search result indicates a slave server to be queried of slave servers, querying, by the first slave server, the slave server to be queried regarding the cached information for materialization;

on condition that the slave server to be queried includes the information materialization, materializing, by the first slave server, the stub-state object using the information for materialization acquired from the slave server to be queried; and

the first slave server synchronizes the copy management information thereon and other copy management information on one of the master server and the slave servers in a checkout process of the first copy file system.

12. The file sharing method according to claim 11 further comprising:

referring to, by the first slave server, a user-settable search policy defining range of the searching; and

searching, by the first slave server, the copy management information for the slayer server to be queried within the search range defined by the search policy.

13. The file sharing method according to claim 11, wherein the copy management information further includes information indicating a location of a slave server;

and

the location information is referred by the first slave server to determine the slave server to be queried.