File sharing system with data mirroring by storage systems

• US 7,281,032 B2
• Filed: 07/13/2001
• Issued: 10/09/2007
• Est. Priority Date: 06/30/2000
• Status: Expired due to Fees

• Current Assignee: Hitachi, Ltd.
• Original Assignee: Hitachi, Ltd.
• Inventors: Kodama, Shoji
• Primary Examiner(s): Caldwell; Andrew
• Assistant Examiner(s): BLAIR, DOUGLAS B
• Application Number:
  US09/905,337
• Publication Number:
  US 20020010762A1
• Time in Patent Office: 2,279 Days
• Field of Search: 709/208, 709217-219, 709/224, 709/214, 707/8, 707/10, 707200-201, 707/204, 711/114
• US Class Current: 709/217
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/10015 Access to distributed or replicated servers, e.g. using brokers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99955 Archiving or backup

Abstract

A system for optimizing data access is provided. The system includes a cluster file system server, a number of cluster file system clients, and a storage system. The storage system includes disk drives organized as paired master disk drive and one or more mirrored disk drives having copies of the data stored on the master disk drive. When a file is needed, the server determines which pair has the needed file, and then determines which disk drive within the pair should be accessed. The determination is made to balance the access load within the pair, and can select to retrieve the file from the drive which is least frequently accessed. The identification of the disk drive to be accessed is then forwarded to the client thereby allowing the client to access the appropriate disk drive to retrieve the file.

Claims

1. A system for optimizing data access, comprising:

a file server capable of communicating with one or more clients; and

a plurality of storage elements organized into pairs for storing a plurality of files, each pair having a master storage element and at least one mirrored storage element, and each mirrored storage element being configured to receive a copy of data stored on the master storage element;

wherein the file server maintains file information on where each of the plurality of files is stored on which pair of storage elements, and maintains access load information regarding each one of the pair of storage elements;

wherein when a client requests access to file information for a requested file, the file server determines a pair of storage elements for accessing the requested file, and returns to the client a first identifier of a storage element within the pair of storage elements that is to be accessed for reading data, and a second identifier of the master storage element of the pair of storage elements that is to be accessed for writing data, the determination being based upon use of storage elements within the system,

wherein the client initiates I/O operations with the pair of storage elements absent the file server,

wherein write operations by the client are made using the first identifier and read operations by the client are made using the second identifier, and

wherein data written to the master storage element is copied to the at least one mirrored storage element within each pair of storage elements.

2. The system according to claim 1 wherein the plurality of storage elements is a plurality of disk drives.

3. The system according to claim 1 wherein the plurality of storage elements are provided in a single storage system.

4. The system according to claim 1 wherein the plurality of storage elements are provided in at least two storage systems.

5. The system according to claim 1 further comprising:

a plurality of host computers;

wherein the file server resides on one of the plurality of host computers; and

wherein the one or more clients reside on remaining ones of the plurality of host computers.

6. The system according to claim 1 further comprising an apparatus configured to synchronize data stored on each pair of storage elements.

7. The system according to claim 1 wherein the master storage element and the at least one storage element within a pair are contained in a single storage system.

8. The system according to claim 1 wherein if it is determined that a mirrored storage element is to be accessed for the requested file and the mirrored storage element which is to be accessed contains a latest copy of data for the requested file stored on the corresponding master storage element, the client directly retrieves the requested file from the mirrored storage element.

9. The system according to claim 8 wherein if it is determined that a mirrored storage element is to be accessed for the requested file and the mirrored storage element which is to be accessed does not contain a latest copy of data for the requested file stored on the corresponding master storage element, the latest copy of data for the requested file stored on the corresponding master storage element is retrieved from the corresponding master storage element and then forwarded to the client.

10. The system according to claim 1 wherein the file information on where each of the plurality of files is stored on which pair of storage elements includes file allocation lists.

11. The system according to claim 1 wherein when determining which of the storage elements within the pair of storage elements having the requested file is to be accessed, consideration is given to ensure that all the storage elements within the pair of storage elements having the requested file are accessed in a substantially balanced manner.

12. The system according to claim 1 wherein upon determining which of the storage elements within the pair of storage elements having the requested file is to be accessed, the file server forwards a file allocation list for the requested file to the client thereby allowing the client to retrieve the requested file from the determined storage element.

13. The system according to claim 12 wherein upon forwarding the file allocation list to the client, the file server updates the access load information to ensure accurate monitoring of access balance of the pairs.

14. A system for optimizing data access comprising:

a first host computer having a file system server, the first host computer capable of communicating with a second host computer having a file system client;

a storage system having a plurality of disk drives organized into pairs for storing a plurality of files, each pair having a master disk drive and at least one mirrored disk drive, each mirrored disk drive being configured to receive a copy of data stored on the master disk drive;

wherein the file system server maintains file information on where each of the plurality of files is stored on each pair of disk drives and further maintains access load information on each pair of disk drives;

wherein when a file system client requests access to file information for a requested file, the file server determines a pair of storage elements for accessing the requested file, and returns to the file system client a first identifier of a storage element within the pair of storage elements that is to be accessed for reading data, and a second identifier of the master storage element of the pair of storage elements that is to be accessed for writing data, the determination being based upon use of storage elements within the system,

wherein the file system client, absent the file system server, initiates I/O operations with the identified disk drive to access the requested file,

wherein write operations by the file system client are made using the first identifier and read operations by the file system client are made using the second identifier, and

wherein data written to the master storage element is copied to the at least one mirrored storage element within each pair of storage elements.

15. The system according to claim 14 wherein the file information on where each of the plurality of files is stored on which pair of disk drives includes file allocation lists.

16. The system according to claim 14 wherein when determining which of the disk drives within the pair of disk drives having the requested file is to be accessed, consideration is given to ensure that all the disk drives within the pair of disk drives having the requested file are accessed in a substantially balanced manner.

17. The system according to claim 14 wherein upon determining which of the disk drives within the pair of disk drives having the requested file is to be accessed, the file system server forwards a file allocation list for the requested file to the file system client thereby allowing the file system client to retrieve the requested file from the determined disk drive.

18. The system according to claim 17 wherein upon forwarding the file allocation list to the file system client, the file system server updates the access load information to ensure accurate monitoring of access balance of the pairs.

19. A system for optimizing data access comprising:

a first host computer having a file system server, the first host computer capable of communicating with a second host computer having a file system client;

a plurality of storage systems, each of the plurality of storage systems having a plurality of disk drives, the plurality of disk drives from the plurality of storage systems being collectively organized into pairs for storing a plurality of files, each pair having a master disk drive and at least one mirrored disk drive, each mirrored disk drive being configured to receive a copy of data stored on the master disk drive;

wherein the file system server maintains file information on where each of the plurality of files is stored on which pair of disk drives and further maintains access load information on each pair of disk drives;

wherein when a file system client requests access to file information for a requested file, the file server determines a pair of storage elements for accessing the requested file, and returns to the file system client a first identifier of a storage element within the pair of storage elements that is to be accessed for reading data, and a second identifier of the master storage element of the pair of storage elements that is to be accessed for writing data, the determination being based upon use of storage elements within the system,

wherein the file system client, absent the file system server, initiates I/O operations with the identified disk drive to access the requested file,

wherein write operations by the file system client are made using the first identifier and read operations by the file system client are made using the second identifier, and

wherein data written to the master storage element is copied to the at least one mirrored storage element within each pair of storage elements.

20. The system according to claim 19 wherein the file information on where each of the plurality of files is stored on which pair of disk drives includes file allocation lists.

21. The system according to claim 19 wherein when determining which of the disk drives within the pair of disk drives having the requested file is to be accessed, consideration is given to ensure that all the disk drives within the pair of disk drives having the requested file are accessed in a substantially balanced manner.

22. The system according to claim 19 wherein upon determining which of the disk drives within the pair of disk drives having the requested file is to be accessed, the file system server forwards a file allocation list for the requested file to the file system client thereby allowing the file system client to retrieve the requested file from the determined disk drive.

23. The system according to claim 22 wherein upon forwarding the file allocation list to the file system client, the file system server updates the access load information to ensure accurate monitoring of access balance of the pairs.

24. The system according to claim 19 wherein if it is determined that a mirrored disk drive is to be accessed for the requested file and the mirrored disk drive which is to be accessed contains a latest copy of data for the requested file stored on the corresponding master disk drive, the file system client directly retrieves the requested file from the mirrored disk drive.

25. The system according to claim 19 wherein if it is determined that a mirrored disk drive is to be accessed for the requested file and the mirrored disk drive which is to be accessed does not contain a latest copy of data for the requested file stored on the corresponding master disk drive, the latest copy of data for the requested file stored on the corresponding master disk drive is retrieved from the corresponding master disk drive and then forwarded to the file system client.

26. A method for optimizing data access comprising:

organizing a plurality of storage elements into pairs for storing a plurality of files, each pair having a master storage element and at least one mirrored storage elements, each mirrored storage element being configured to receive a copy of data stored on the master storage element;

maintaining file information on where each of the plurality of files is stored on each pair of storage elements;

maintaining access load information on each one of the pair of storage elements;

upon receiving a request for a requested file, determining a pair of storage elements for accessing the requested file,

returning file access information including a first identifier of a storage element within the pair of storage elements that is to be accessed for reading data, and a second identifier of the master storage of the pair of storage elements that is to be accessed for writing data, the first identifier being selected based upon use of the storage elements,

wherein said file access information is returned to a client thereby allowing the client to retrieve the requested file absent a file server,

wherein write operations by the client are made using the first identifier and read operations by the client are made using the second identifier, and

wherein data written to the master storage element is copied to the at least one mirrored storage element within each pair of storage elements.

27. The method of claim 26 further comprising upon returning file access information to the client, updating the access load information to ensure accurate monitoring of access balance of the pairs.

28. The method according to claim 26 wherein the plurality of storage elements are stored on a single storage system.

29. The method according to claim 26 wherein the plurality of storage elements are stored in at least one storage system.

30. The method according to claim 29 further comprising upon determining that a mirrored storage element is to be accessed for the requested file and the mirrored storage element which is to be accessed contains a latest copy of data for the requested file stored on the corresponding master storage element, retrieving the requested file from the mirrored storage element directly.

31. The method according to claim 29 further comprising upon determining that a mirrored storage element is to be accessed for the requested file and the mirrored storage element which is to be accessed does not contain a latest copy of data for the requested file stored on the corresponding master storage element, retrieving the latest copy of data for the requested file stored on the corresponding master storage element from the corresponding master storage element.

32. The method according to claim 26 wherein when determining which of the storage elements within the pair of storage elements having the requested file is to be accessed, consideration is given to ensure that all the storage elements within the pair of storage elements having the requested file are substantially accessed in a balanced manner.

33. The method according to claim 26 wherein the plurality of storage elements includes a plurality of disk drives.

34. A method for optimizing data access between a file server and at least one or more clients comprising:

organizing a plurality of disk drives into pairs for storing a plurality of files, each pair having a master disk drive and at least one mirrored disk drive, and each mirrored disk drive being configured to receive a copy of data stored on the master drive;

maintaining file information on where each of the plurality of files is stored on which pair of disk drives;

maintaining access load information on each pair of disk drives, wherein the file information and the access load information are maintained at the file server;

upon receiving a request for a requested file, determining a pair of storage elements for accessing the requested file based upon a usage level of its corresponding storage elements, and

forwarding to the client a first identifier of a storage element within the pair of storage elements that is to be accessed for reading data, and a second identifier of the master storage within the pair of storage elements for writing data thereby allowing the client to initiate I/O operations, absent the file server, to access the requested file.