Distributed metadata cache
Apparatuses, methods, and other embodiments associated with distributed metadata caching are described. According to one embodiment, a method includes intercepting a metadata request intended for an NFS server and communicating with a member of the set of peers to satisfy the metadata request from a distributed metadata cache maintained by the set of peers. The metadata request will have originated from a member of a set of peers participating in peer-to-peer distributed metadata caching.
1. A method, comprising:
intercepting a metadata request, in a computer, intended for a network file system (NFS) server, wherein the metadata request originated from a member of a set of peers participating in peer-to-peer distributed metadata caching;
controlling the computer to identify a member of the set of peers to communicate with to satisfy the metadata request by accessing a per file data structure that stores information correlating files to one or more members of the set of peers, wherein the per file data structure stores a file identifier and a related client identifier, and wherein the related client identifier uniquely identifies the member of the set of peers that has a most up-to-date metadata for a local copy of a file specified by the metadata request; and
communicating with the member of the set of peers to satisfy the metadata request from a distributed metadata cache maintained by the set of peers.
2. The method of claim 1 , wherein communicating with the member of the set of peers to satisfy the metadata request comprises one or more of, satisfying a request to acquire metadata, and satisfying a request to update metadata.
3. The method of claim 1 , comprising identifying the member of the set of peers to communicate with to satisfy the metadata request based, at least in part, on a file identifier associated with the metadata request.
4. The method of claim 3 , wherein identifying the member of the set of peers to communicate with comprises applying a hash function to the file identifier associated with the metadata request to acquire data from which a peer identifier can be acquired.
5. The method of claim 1 , comprising controlling the computer to store one or more of, current metadata for a local copy of a file stored at a different member of the set of peers, and information to identify a member of the set of peers that has current metadata for a local copy of a file stored at a different member of the set of peers.
6. A non-transitory computer-readable medium storing computer executable instructions that when executed by a computer cause the computer to perform a method, the method comprising:
intercepting a metadata request, in the computer, intended for a network file system (NFS) server, wherein the metadata request originated from a member of a set of peers participating in peer-to-peer distributed metadata caching;
controlling the computer to identify a member of the set of peers to communicate with to satisfy the metadata request by accessing a per file data structure that stores information correlating files to one or more members of the set of peers, wherein the per file data structure stores a file identifier and a related client identifier, and wherein the related client identifier uniquely identifies the member of the set of peers that has a most up-to-date metadata for a local copy of a file specified by the metadata request; and
communicating with the member of the set of peers to satisfy the metadata request from a distributed metadata cache maintained by the set of peers.
7. The non-transitory computer-readable medium of claim 6 , wherein communicating with the member of the set of peers to satisfy the metadata request comprises one or more of, satisfying a request to acquire metadata, and satisfying a request to update metadata.
8. The non-transitory computer-readable medium of claim 6 , comprising identifying the member of the set of peers to communicate with to satisfy the metadata request based, at least in part, on a file identifier associated with the metadata request.
9. The non-transitory computer-readable medium of claim 8 , wherein identifying the member of the set of peers to communicate with comprises applying a hash function to the file identifier associated with the metadata request to acquire data from which a peer identifier can be acquired.
10. The non-transitory computer-readable medium of claim 6 , comprising controlling the computer to store one or more of, current metadata for a local copy of a file stored at a different member of the set of peers, and information to identify a member of the set of peers that has current metadata for a local copy of a file stored at a different member of the set of peers.
11. The system of claim 6 , wherein communicating with the member of the set of peers to satisfy the metadata request comprises one or more of, satisfying a request to acquire metadata, and satisfying a request to update metadata.
12. The system of claim 6 , comprising identifying the member of the set of peers to communicate with to satisfy the metadata request based, at least in part, on a file identifier associated with the metadata request.
13. The system of claim 12 , wherein identifying the member of the set of peers to communicate with comprises applying a hash function to the file identifier associated with the metadata request to acquire data from which a peer identifier can be acquired.
14. The system of claim 6 , comprising controlling the computer to store one or more of, current metadata for a local copy of a file stored at a different member of the set of peers, and information to identify a member of the set of peers that has current metadata for a local copy of a file stored at a different member of the set of peers.
15. A system, comprising:
an intercept logic configured to intercept a metadata request, in a computer, intended for a network file system (NFS) server, wherein the metadata request originated from a member of a set of peers participating in peer-to-peer distributed metadata caching; and
a communication logic configured to:
control the computer to identify a member of the set of peers to communicate with to satisfy the metadata request by accessing a per file data structure that stores information correlating files to one or more members of the set of peers, wherein the per file data structure stores a file identifier and a related client identifier, and wherein the related client identifier uniquely identifies the member of the set of peers that has a most up-to-date metadata for a local copy of a file specified by the metadata request; and
communicate with the member of the set of peers to satisfy the metadata request from a distributed metadata cache maintained by the set of peers.