Optimizing wide area network (WAN) traffic by providing home site deduplication information to a cache site

A system for optimizing wide-area network traffic comprising:

• a remote cluster file system including a home storage server;
• a local cache cluster including a local cache manager in communication with a local cache;
• and
• a wide area network providing communication between the remote cluster file system and the local cache cluster,
• wherein the local cache manager is configured:
• to fetch deduplication information from the home storage server to build a repository including duplicate peer file sets and a table file, each duplicate peer file set being a distinct group and each including one or more peer files and at least one duplicate peer file set including two or more peer files, the one or more peer files in a single duplicate file set being copies of the same file, the table file comprising a listing of duplicate peer file sets and an indication as to whether a cache copy of the one or more peer files exists on a local cache;
• to receive a request to access a target file;
• in response to the request, to refer to the repository to determine whether the target file corresponds with a peer file of the one or more peer files included in the duplicate peer file sets and whether a cache copy of the peer file of the one or more peer files exists on the local cache;
• to create a local copy of the peer file from the cache copy of the peer file of the one or more peer files, if a determination is made that the target file corresponds with the peer file of the one or more peer files included in the duplicate peer file sets and that the cache copy of the peer file of the one or more peer files exists on the local cache;
• and
• wherein the local cache manager is configured to interact with a client application and to make a copy of a corresponding peer file of the one or more peer files in the repository, giving the copy of the corresponding peer file a file name matching that of the target file and to provide access to the copied corresponding peer file to the client application, if a determination is made that the target file corresponds with the peer file of the one or more peer files included in the duplicate peer file sets and no cache copy of the peer file of the one or more peer files exists on the local cache.

A physical computer storage medium comprising a computer program product for optimizing wide-area network traffic, the physical computer storage medium comprising:

• computer code for fetching deduplication information from a home site to build a repository including duplicate peer file sets and a table file, each duplicate peer file set being a distinct group and each including one or more peer files and at least one duplicate peer file set including two or more peer files, the one or more peer files in a single duplicate file set being copies of the same file, the table file comprising a listing of the duplicate peer file sets and an indication as to whether a cache copy of the one or more peer files exists on a local cache;
• computer code for receiving a request to access a target file;
• computer code for, in response to the request, referring to the repository to determine whether the target file corresponds with a peer file of the one or more peer files included in the duplicate peer file sets and whether a cache copy of the peer file of the one or more peer files exists on the local cache;
• computer code for creating a local copy of the peer file from the cache copy of the peer file of the one or more peer files, if a determination is made that the target file corresponds with the peer file of the one or more peer files included in the duplicate peer file sets and that the cache copy of the peer file of the one or more peer files exists on the local cache;
• and
• computer code for making a copy of a corresponding peer file of the one or more peer files in the repository, giving the copy of the corresponding peer file a file name matching that of the target file, and providing access to the copied corresponding peer file to a client application, if a determination is made that the target file corresponds with the peer file of the one or more peer files included in the duplicate peer file sets and no cache copy of the peer file of the one or more peer files exists on the local cache.