Abstract: A snapshot is received at the MDS from a backup storage location at a point-in-time subsequent to taking the snapshot, wherein the MDS comprises a map identifying an object store location for a plurality of data objects and a queue identifying individuals of the plurality of data objects that only reside on the transient object store. Individuals of the plurality of data objects identified in the queue are located on the distributed file system. The map is updated to identify the object store location for the individuals of the plurality of data objects identified in the queue.

Abstract: Systems and methods for migrating a metadata server are provided. A source metadata server is transferred to a new metadata server. During the migration process, only one of the metadata servers is active. Over time, data is transferred to the new metadata server until the new metadata server is fully synchronized with the source metadata server. Then metadata service process is terminated on the source metadata server and started on the new metadata server. The new metadata server transitions to read/write and normal file system operations are resumed at the new metadata server.

Abstract: A metadata server determines (“MDS”) a state associated with each of a plurality of object stores, wherein each object store is at an object store location. The state associated with each of the plurality of object stores is stored on the MDS. The MDS processes a client request for the object store location associated with one of the plurality of object stores in accordance with the state.

Abstract: A snapshot is received at the MDS from a backup storage location at a point-in-time subsequent to taking the snapshot, wherein the MDS comprises a map identifying an object store location for a plurality of data objects and a queue identifying individuals of the plurality of data objects that only reside on the transient object store. Individuals of the plurality of data objects identified in the queue are located on the distributed file system. The map is updated to identify the object store location for the individuals of the plurality of data objects identified in the queue.

Abstract: A partitioned Bloom filter is disclosed. In various embodiments, a representation of an item is received. The representation is used to determine a partition with which the item is associated. A partition-specific Bloom filter is used to determine at least in part whether the item may be an element of a set with which the partition is associated.

Abstract: A partitioned Bloom filter is disclosed. In various embodiments, a representation of an item is received. The representation is used to determine a partition with which the item is associated. A partition-specific Bloom filter is used to determine at least in part whether the item may be an element of a set with which the partition is associated.

Abstract: A snapshot is received at the MDS from a backup storage location at a point-in-time subsequent to taking the snapshot, wherein the MDS comprises a map identifying an object store location for a plurality of data objects and a queue identifying individuals of the plurality of data objects that only reside on the transient object store. Individuals of the plurality of data objects identified in the queue are located on the distributed file system. The map is updated to identify the object store location for the individuals of the plurality of data objects identified in the queue.

Abstract: An indication to retire a computer readable storage medium from the site cache is received. A plurality of data objects stored on the computer readable storage medium are replicated to at least one target computer readable storage medium in the site cache. The computer readable storage medium is removed from the site cache.

Abstract: A failing computer readable storage medium comprising a portion of a transient object store is detected at a site cache. The site cache associated with the transient object store is set to an error read only state. The failing computer readable storage medium is replaced with an operational computer readable storage medium.

Abstract: A method, article of manufacture, and apparatus for providing a site cache in a distributed file system is discussed. Data objects may be written to a site cache prior to writing them to an authoritative object store. This provides performance benefits when a client writing the data has a better connection to the site cache than to the authoritative object store.