Emc Corporation

Emc IP Holding Company LLC

Darrick Peter Lew

Jeffrey Douglas Hughes

Shantanu Sardesai

Pankaj Datta

Harsha Mahuli

Dmitri Chmelev

Bharat Sharma

Steven Lockhart

Nathan Koppes

Tyler Oshiro

Embodiments of the present disclosure relate to creating and maintaining a separate compliance store within a domain. Files that are committed within the domain are linked to the compliance store using canonical links. Canonical links are guaranteed to be non-erasable and non-rewritable for the duration of their retention period. When using a source cluster of nodes as a primary production cluster and a target cluster of nodes as a backup cluster, file data and canonical links can be synced from the source cluster to the target cluster to maintain compliance with data retention rules. Non-canonical links can be modified as necessary to ensure snapshot consistency when failover or failback occurs between a source cluster of nodes and a target cluster of nodes.

Compliance namespace separation to achieve snapshot based consistency across failover-failback while maintaining data retention regulation compliance

Data sets are synchronized between two or more clusters of nodes that support different versions of files (e.g., stub files) within a distributed file storage system. Moreover, the distributed file storage system employs a tiered cloud storage architecture. In one aspect, for stub files having versions that are not commonly supported by the two or more clusters, an application protocol interface (API) is utilized that employs a deep-copy process wherein cloud-backed data referenced by a stub file is retrieved from a cloud storage (e.g., public cloud) and sent by a primary cluster to one or more secondary clusters. Moreover, the API can determine an optimal synchronization process on a per-file basis.

Systems and methods for bi-directional replication of cloud tiered data across incompatible clusters

The various embodiments are directed towards enabling access to snapshotted versions of files in a distributed file system. The distributed file system may receive a file system operation request that includes at least an entry-ID and a snapshot-ID. File system operation requests may include operations such as reads, writes, moves, or the like. Directory entries that match the entry-ID and the snapshot-ID may be looked up and retrieved from the distributed file system. The retrieved directory entries may also be arranged to include a generation-ID. In the directory entry matching the snapshot-ID, if the generation-ID and the snapshot-ID are different, another directory entry may be looked up from the distributed file system using the entry-ID and the generation-ID. Then the initially requested file system operation may be performed on the retrieved directory entry.

Dynamic network access of snapshotted versions of a clustered file system

Embodiments are directed towards replication and restoration of file system objects stored on file system clusters. Modified file system objects may be tracked using a snapshot tracking file corresponding to a snapshot. The snapshot tracking file may be employed to generate a snapshot change set. The snapshot change set may be used to restore a file system cluster to a version corresponding to a snapshot by reversing the changes in the snapshot change set. Also, a snapshot change set may be used to replicate file system objects by committing the included changes on another file system cluster. If a primary file system cluster becomes unavailable the file system may failover to the secondary file system cluster. If the primary file system cluster later becomes available it may be restored by using snapshot change sets generated on the secondary file system cluster.

Replication and restoration

Growing community of inventors

Seattle, WA, United States of America

Darrick Peter Lew