File storage system, cache appliance, and method

• US 20100094806A1
• Filed: 09/18/2008
• Published: 04/15/2010
• Est. Priority Date: 09/18/2008
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Arraid, Inc. (Reactive Group)
• Inventors: Apostolides, Vasilis J., Bertschi, Jason, Hosteny, Joseph Nevi IV, Kazar, Michael L., Boyles, John R.
• Granted Patent: US 9,323,681 B2
• Time in Patent Office: Days
• US Class Current: 707/637
• CPC Class Codes: G06F 12/0804 with main memory updating G06F12/0806 takes precedence
• CPC Class Codes: G06F 12/0815 Cache consistency protocols
• CPC Class Codes: G06F 12/0866 for peripheral storage systems, e.g. disk cache
• CPC Class Codes: G06F 16/172 Caching, prefetching or hoarding of files
• CPC Class Codes: G06F 16/1767 Concurrency control, e.g. optimistic or pessimistic approaches
• CPC Class Codes: G06F 16/183 Provision of network file services by network file servers, e.g. by using NFS, CIFS network file access protocols H04L67/1097
• CPC Class Codes: G06F 2212/264 Remote server
• CPC Class Codes: G06F 2212/284 being distributed
• CPC Class Codes: G06F 2212/314 In storage network, e.g. network attached cache
• CPC Class Codes: G06F 2212/463 File

Abstract

A file storage system for storing data of a file received from a client includes a back-end file server in which the data is stored. The system includes a cache appliance in communication with the file server, such that the appliance stores portions of the data or attributes of the file, and uses the stored data or attributes to process file system requests received from the client, and which reads and writes data and attributes to the back-end file server independently. A system for responding to a file system protocol request in regard to a back-end server includes a token server. The system includes a plurality of cache appliances in communication with the token server, each of which receives tokens from the token server to synchronize access to data and attributes caches of the cache appliances, and reading and writing data and attributes to the back-end servers when tokens are revoked, the cache appliance having persistent storage in which data are stored, and the token server having persistent storage in which tokens are stored. A storage system includes a plurality of backend servers. The system includes a token server which grants permission to read and write file attributes and data system, and includes a plurality of cache appliances in communication with at least one of the backend servers and the token server for processing an incoming NFS request to the one backend server. Each cache appliance comprises an NFS server which converts incoming NFS requests into cache manager operations; a token client module in communication with the token server having a cache of tokens obtained from the token server; a cache manager that caches data and attributes and uses tokens from the token client module to ensure that the cached data or attributes are the most recent data or attributes, and an NFS client which sends outgoing NFS requests to the back-end file server. Methods for storing data of a file received from a client.

Claims

1. A file storage system for storing data of a file received from a client comprising:

a back-end file server in which the data is stored; and

a cache appliance in communication with the file server, such that the appliance stores portions of the data or attributes of the file, and uses the stored data or attributes to process file system requests received from the client, and which reads and writes data and attributes to the back-end file server independently.

13. A system as in claim 1 wherein a protocol used to communicate with the back-end file server is NFS.

14. A system as described in claim 1 wherein read data tokens and read attribute tokens are leased and are automatically revoked when their lease expires.

2. A system for responding to a file system protocol request in regard to a back-end server comprising:

a token server; and

a plurality of cache appliances in communication with the token server, each of which receives tokens from the token server to synchronize access to data and attributes caches of the cache appliances, and reading and writing data and attributes to the back-end servers when tokens are revoked, the cache appliance having persistent storage in which data and attributes are stored, and the token server having persistent storage in which tokens are stored.

3. A system as described in claim 2 wherein the token server produces a write data token to write data to the back-end server by the cache appliance, and ensures that no two write data tokens are granted over a single byte of any same file.

4. A system as described in claim 3 wherein the token server produces write attribute tokens for files stored in the back-end server and ensures that no two write attribute tokens are granted for the same file, and wherein the token server revokes an already granted write data or write attribute token by recalling the granted write data or write attribute token if a new write data or write attribute token is requested from the token server and is incompatible with the already granted write attribute or write data token.

5. A system as described in claim 4 including a spare token server to which the write data tokens and write attribute tokens are mirrored.

6. A system as described in claim 4 including another cache appliance where modified data and modified attributes are mirrored and persistently stored.

7. A system as described in claim 4 wherein the token server produces read data tokens and read attribute tokens that have a per token expiration time which allows them to be unilaterally revoked by the token server after the expiration time has passed.

8. A system as described in claim 4 wherein the token server produces ownership tokens for the cache appliance for the file'"'"'s attributes and data, and data and attributes associated with the ownership tokens are retrieved directly from the cache appliance having the ownership tokens.

9. A system as described in claim 4 wherein the token server produces a ghost token persisting for a time period during which the token server indicates a last time in which a write data or attribute token was held for the file.

10. A system as described in claim 4 where the token server produces batch tokens to reduce communication with the token server for shared files.

11. A system as described in claim 5 wherein dual logs, or a circular log, is used to store tokens persistently at the token server.

12. A system as described in claim 2 wherein a token server is chosen by a function including a file handle as a parameter, which maps the file handle to a plurality of token servers, with a plurality of token servers in communication with the plurality of cache appliances.

15. A method for storing data of a file received from a client comprising the steps of:

storing portions of the data or attributes of the file in a cache appliance;

using the stored data or attributes to process file system requests received from the client;

reading and writing the data and the attributes to the back-end file server independently; and

storing the data and the attributes in a back-end file server in communication with the cache appliance.

16. A method for responding to a file system protocol request in regard to a back-end server comprising the steps of:

receiving tokens from a token server in communication with a plurality of cache appliances;

synchronizing access to data caches and attributes caches of the cache appliances with the tokens;

reading and writing file data and attributes in the caches of the cache appliances holding tokens; and

writing data in the data caches and attributes in the attributes caches to a back-end server when the tokens are revoked, the cache appliance having persistent storage in which the data and attributes are stored, and the token server having persistent storage in which tokens are stored.

17. A method as described in claim 16 including the step of the production of a write data token by the token server for files stored in the back-end server, the token server ensuring that no two write data tokens are granted over a single byte of any same file.

18. A method as described in claim 17 including the step of the production of write attribute tokens by the token server for files stored in the back-end server, the token server ensuring that no two write attribute tokens are granted for the same file.

19. A method as described in claim 18 including the step of the token server revoking an already granted write data or write attribute token by recalling the granted write data or write attribute token if a new write data or write attribute token is requested from the token server and is incompatible with the already granted write attribute or write data token.

20. A method as described in claim 19 including the step of mirroring in a spare token server the write data tokens and write attribute tokens.

21. A method as described in claim 19 including the step of persistently storing in another cache appliance modified data and modified attributes.

22. A method as described in claim 19 including the step of producing with the token server read data tokens and read attribute tokens that have a per token expiration time which allows them to be unilaterally revoked by the token server after the expiration time has passed.

23. A method as described in claim 19 including the step of producing with the token server ownership tokens for the cache appliance for the file'"'"'s attributes and data.

24. A method as described in claim 23 including the step of retrieving directly from the cache appliance data and attributes of files for which the cache appliance holds an ownership token.

25. A method as described in claim 19 including the step of the token server producing a ghost token persisting for a time period during which the token server indicates a last time in which a write data or attribute token was held for the file.

26. A method as described in claim 19 including the step of the token server producing batch tokens to reduce communication with the token server for shared files.

27. A method as described in claim 16 including the step of choosing the token server by a function which includes a file handle as a parameter, which maps the file handle to a plurality of token servers.

28. A method as described in claim 16 including the steps of leasing the read data tokens and read attribute tokens and automatically revoking them when their lease expires.

29. A storage system comprising:

a plurality of backend servers;

a token server which grants permission to read and write file attributes and data;

a plurality of cache appliances in communication with at least one of the backend servers and the token server for processing an incoming NFS request to the one backend server, each cache appliance comprising;

an NFS server converts incoming NFS requests into cache manager operations;

a token client module in communication with the token server having a cache of tokens obtained from the token server;

a cache manager that caches data and attributes and uses tokens from the token client module to ensure that the cached data or attributes are the most recent data or attributes, and

an NFS client which sends outgoing NFS requests to the back-end file server.

30. A system as described in claim 29 including a cache file system in communication with the cache manager which stores copies of file attributes, file data and directory listing information.

31. A system as described in claim 30 including a C2C interface which accesses data and attributes in other cache appliances.

32. A method for storing data comprising the steps of:

processing an incoming NFS request with a cache appliance of a plurality of cache appliances communicating with at least one of a plurality of backend servers, each cache appliance comprising;

an NFS server which converts incoming NFS requests into cache manager operations;

a token client module in communication with the token server having a cache of tokens obtained from the token server;

a cache manager that caches data and attributes and uses tokens from the token client module to ensure that the cached data or attributes are the most recent data or attributes, and

an NFS client which sends outgoing NFS requests to the back-end file server; and

obtaining permission by the cache appliance from a token server to read and write file attributes and data.

33. A method as described in claim 32 including the step of storing copies of file attributes and file data and directory listing information in a cache file system in communication with the cache manager.

34. A method as described in claim 33 including the step of accessing data and attributes in other cache appliances using a C2C interface of the cache appliance.