Abstract: A method and apparatus for different embodiments of probabilistic summary data structure based encoding for garbage collection are described. In one embodiment, a method comprises generating a probabilistic summary data structure that represents active blocks of data within a storage device based on identifications of the active blocks or the data within the active blocks. The method also includes performing garbage collection of at least a portion of the storage device based on the probabilistic summary data structure.

Abstract: A system and method are disclosed for providing efficient data storage. A plurality of data segments is received in a data stream. The system determines whether a data segment has been stored previously in a low latency memory. In the event that the data segment is determined to have been stored previously, an identifier for the previously stored data segment is returned.

Abstract: A network file system-based data storage system that converts random I/O requests into a piecewise sequential data structure to facilitate variable length data segment redundancy identification and elimination. For one embodiment of the invention a stateless network file system is employed. For one such embodiment, that provides multiple-client access to stored data, multiple Writes are buffered and then broken into variable length data segments. Redundant segment elimination is then effected. One embodiment of the invention allows sharing of the variable length data segments among files.

Abstract: A method and apparatus for different embodiments of probabilistic summary data structure based encoding for garbage collection are described. In one embodiment, a method comprises generating a probabilistic summary data structure that represents active blocks of data within a storage device based on identifications of the active blocks or the data within the active blocks. The method also includes performing garbage collection of at least a portion of the storage device based on the probabilistic summary data structure.

Abstract: A network file system-based data storage system that converts random I/O requests into a piecewise sequential data structure to facilitate variable length data segment redundancy identification and elimination. For one embodiment of the invention a stateless network file system is employed. For one such embodiment, that provides multiple-client access to stored data, multiple Writes are buffered and then broken into variable length data segments. Redundant segment elimination is then effected. One embodiment of the invention allows sharing of the variable length data segments among files.

Abstract: A method for storing data is disclosed. The method comprises receiving a data stream comprising a plurality of data segments and preliminarily checking in a memory having a relatively low latency whether one of the plurality of data segments has been stored previously. The method further comprises in the event that the preliminary check does not conclusively determine whether the data segment has been stored previously, limiting checking in a memory having a relatively high latency to conclusively determine whether the data segment has been previously stored, and in the event that checking is limited or in the event that the check in the memory having relatively high latency conclusively determines the data segment has not been previously stored, storing the data segment.

Abstract: Verifying a file in a system with duplicate segment elimination is disclosed. A data file is segmented into a plurality of distinct data segments, and a checksum is computed for each of the plurality of distinct data segments. A constructed data file checksum is constructed from the checksums of each of the plurality of distinct data segments, and, it is determined if a checksum of the data file is the same as the constructed data file checksum.

Abstract: A system and method are disclosed for processing a data stream. A data segment is received. It is determined whether the data segment has been previously stored. In the event that the data segment is determined not to have been previously stored, a unique identifier for specifying the data segment in a representation of the data stream is generated.

Abstract: A method and apparatus for storing composite data streams. According to one embodiment of the invention, a composite data stream is stored so that it may be restored. The storing of the composite data stream includes decomposing the composite data stream into a plurality of constituent data streams, segmenting at least one of the plurality of constituent data streams, and discarding those of the segments resulting from the segmenting which are determined to have been stored previously.

Abstract: A method and apparatus for backing up a computer system is disclosed. A backup is provided of a data stored on a first computer by the first computer generating a dump stream. The dump stream is transmitted to a second computer, the second computer uses the dump stream to maintain a file system copy of the data. One or more snapshots of the file system of the second computer are stored on a backup computer, the snapshots providing a backup of the data stored on the first computer.

Abstract: Selecting a segment boundary comprises determining a segmentation window is determined and computing a plurality of values associated with candidate boundaries within the segmentation window. Selecting a segment boundary further comprises selecting one of the candidate boundaries based at least in part on a comparison between two or more of the computed values and determining a boundary within the segmentation window.

Abstract: A system and method for managing a plurality of snapshots as provided. A set of metadata describing a file system is contained within the file system so that a snapshot of the file system includes the associated metadata. Backup client file systems are restored to a backup server using conventional dump and restore techniques. The backup server then utilizes a user-defined snapshot management schedule to manage the set of backups associated with the backup server. Such management of snapshots can include deletion of snapshots based upon a variety of parameters including the time stamp.

Abstract: A system and method for coalescing a plurality of snapshots into a single destination volume file system. Each snapshot and its replicated file system is replicated in a directory associated with the destination file system in which the coalesced snapshots are stored. The destination file system can be associated with a backup server and information from each of the plurality of snapshots can be transferred to the backup server by an respective backup clients over a network. Information can be passed to the backup server from the backup client(s) using a dump and restore process.

Abstract: Cluster storage is disclosed. A data stream or a data block is received. The data stream or the data block is broken into segments. For each segment, a cluster node is selected, and in the event that a similar segment to the segment is identified that is already managed by the selected cluster node, a reference to the similar segment and a delta between the similar segment and the segment is caused to be stored on the selected cluster node.

Abstract: A method and apparatus for different embodiments of incremental garbage collection of data in a secondary storage. In one embodiment, a method comprises locating blocks of data in a log that are referenced and within a range at a tail of the log. The method also includes copying the blocks of data that are referenced and within the range to an unallocated segment of the log.

Abstract: Cluster storage is disclosed. A data stream or a data block is received. The data stream or the data block is broken into segments. For each segment, a cluster node is selected, and a portion of the segment smaller than the segment is identified that is a duplicate of a portion of a segment already managed by the cluster node.

Abstract: Seeding replication is disclosed. One or more but not all files stored on a deduplicated storage system are selected to be replicated. One or more segments referred to by the selected one or more but not all files are determined. A data structure is created that is used to indicate that at least the one or more segments are to be replicated. In the event that an indication based at least in part on the data structure indicates that a candidate segment stored on the deduplicating storage system is to be replicated, the candidate segment is replicated.

Abstract: Cluster storage is disclosed. A data stream or a data block is received. The data stream or the data block is broken into segments. For each segment, a cluster node is selected, and a portion of the segment smaller than the segment is identified that is a duplicate of a portion of a segment already managed by the cluster node.

Abstract: Transmitting or storing subsegments is disclosed. A data stream or a data block is received and broken into a plurality of segments. For at least one segment, the segment is broken into a plurality of subsegments. A previously stored or transmitted segment similar to the at least one segment is identified. A fingerprint is computed for at least one subsegment. And, using the fingerprint for the at least one subsegment, determining whether the at least one subsegment is identical to a subsegment of the previously stored or transmitted segment without directly comparing the content of the at leas one subsegment with the content of the subsegment of the previously stored or transmitted segment.

Abstract: A system and method are disclosed for providing efficient data storage. A data stream comprising a plurality of data segments is received. The system determines whether one of the plurality of data segments has been stored previously using a summary in a low latency memory; in the event that the data segment is determined not to have been stored previously, assigning an identifier to the data segment.