Reliable hierarchical storage management with data synchronization

A method comprising:

• hosting a file system representing a replica of file systems hosted on a plurality of nodes, each of the plurality of nodes storing data on a first storage medium;
• storing data associated with the file system on a second storage medium, the second storage medium being remote from the plurality of nodes and slower to process data stored thereon than the first storage medium;
• receiving a file from one of the plurality of nodes when the file remains un-accessed by the plurality of nodes for a predetermined period of time;
• storing the file on the second storage medium, the plurality of nodes storing a pointer to the file and not the data content of the file on the first storage medium;
• providing the plurality of nodes read and write access to the file stored on the second storage medium using the pointer on the first storage medium;
• allowing concurrent opens of the file across the plurality of nodes without providing the data content of the file to the first storage medium of each node of the plurality of nodes, wherein allowing concurrent opens of the file comprises synchronizing the file to the file systems on the plurality of nodes prior to completion of opening the file on one of the plurality of nodes.

A hierarchical storage management system comprising:

• a processor;
• and
• machine readable instructions, stored on a tangible machine readable medium, when executed by the processor, configure the processor to:
• host a file system representing a replica of file systems hosted on a plurality of nodes, each of the plurality of nodes storing data on a first storage medium;
• store data associated with the file system on a second storage medium that is remote from the plurality of nodes and slower to process data stored thereon than the first storage medium;
• receive a file from one of the plurality of nodes when the file remains un-accessed by the plurality of nodes for a predetermined period of time;
• store the file on the second storage medium, a pointer to the file and not the data content of the file being stored on the first storage medium on the plurality of nodes;
• provide the plurality of nodes read and write access to the file stored on the second storage medium using the pointer on the first storage medium;
• allow concurrent opens of the file across the plurality of nodes without providing the data content of the file to the first storage medium of each node of the plurality of nodes, wherein the instructions further configure the processor to, upon a node requesting an open of the file, synchronize metadata of the file to the file systems on the plurality of nodes prior to completion of opening the file on the node requesting the open of the file.