Caringo, Inc.

Russell Turpin

Paul R M Carpentier

Jonathan Ring

David Yoakley

Andrew David Klager

Don Baker

Aaron Pierce

Glen Olsen

Paul Rm Carpentier

A client application is provided with a view of a subset of objects within an object store. A REST protocol, REST interface, or API may be used to access the objects. A coupling server may include a filter and a generator. The generator may automatically generate object metadata for newly created files that matches a filter. A client application may mount a drive on a computer based on metadata objects, such as providing a filter on object metadata to view files a user wants to access; or editing, reading, and updating object metadata as a file.

Mounting dynamic endpoints

Digital objects are stored and accessed within a fixed content storage cluster by using a page mapping table and a pages index. A stream is read from the cluster by using a portion of its unique identifier as a key into the page mapping table. The page mapping table indicates a node holding a pages index indicating where the stream is stored. A stream is written by storing the stream on any suitable node and then updating a pages index stored within the cluster responsible for knowing the location of digital objects having unique identifiers that fall within a particular address range. The cluster recovers from a node failure by first replicating streams from the failed node and reallocating a page mapping table to create a new pages index. The remaining nodes send records of the unique identifiers corresponding to objects they hold to the new pages index.

Two level addressing in storage clusters

A cluster receives a request to store an object using replication or erasure coding. The cluster writes the object using erasure coding. A manifest is written that includes an indication of erasure coding and a unique identifier for each segment. The cluster returns a unique identifier of the manifest. The cluster receives a request from a client that includes a unique identifier. The cluster determines whether the object has been stored using replication or erasure coding. If using erasure coding, the method reads a manifest. The method identifies segments within the cluster using unique segment identifiers of the manifest. Using these unique segment identifiers, the method reconstructs the object. A persistent storage area of another disk is scanned to find a unique identifier of a failed disk. If using erasure coding, a missing segment previously stored on the disk is identified. The method locates other segments. Missing segments are regenerated.

Erasure coding and replication in storage clusters

A method writes a data block from a client application onto object storage. Object storage receives a write request with a data block over an API from a coupling server, the data block being an update to a target file stored within object storage. Next, object storage receives a completion request from the coupling server for the target file. Then, object storage receives a final manifest from the coupling server, the final manifest includes for each data block of the target file an identifier for said data block, a start location, an end location or a size, and a unique identifier for the data block; the manifest has the data blocks ordered by positions in the target file. Finally, object storage assembles of new version of the target file using the final manifest, the received data block and other data blocks of the target file, and writes a new version into object storage.

Multi-part upload

Digital objects are stored and accessed within a fixed content storage cluster by using a page mapping table and a pages index. A stream is read from the cluster by using a portion of its unique identifier as a key into the page mapping table. The page mapping table indicates a node holding a pages index indicating where the stream is stored. A stream is written to the cluster by storing the stream on any suitable node and then updating a pages index stored within the cluster. The cluster recovers from a node failure by first replicating streams from the failed node and reallocating a page mapping table to create a new pages index. The remaining nodes send records of the unique identifiers corresponding to objects they hold to the new pages index. A node is added to the cluster by reallocating a page mapping table.

Digital objects within a fixed-content storage cluster use a page mapping table and a hash-to-UID table to store a representation of each object. For each object stored within the cluster, a record in the hash-to-UID table stores the object's hash value and its unique identifier (or portions thereof). To detect a duplicate of an object, a portion of its hash value is used as a key into the page mapping table. The page mapping table indicates a node holding a hash-to-UID table indicating currently stored objects in a particular page range. Finding the same hash value but with a different unique identifier in the table indicates that a duplicate of an object exists. Portions of the hash value and unique identifier may be used in the hash-to-UID table. Unneeded duplicate objects are deleted by copying their metadata to a manifest and then redirecting unique identifiers to point at the manifest.

Elimination of duplicate objects in storage clusters

Digital objects within a fixed-content storage cluster use a page mapping table and a hash-to-UID table to store a representation of each object. For each object stored within the cluster, a record in the hash-to-UID table stores the object's hash value and its unique identifier (or portions thereof). To detect a duplicate of an object, a portion of its hash value is used as a key into the page mapping table. The page mapping table indicates a node holding a hash-to-UID table indicating currently stored objects in a particular page range. Finding the same hash value but with a different unique identifier in the table indicates that a duplicate of an object exists. Portions of the hash value and unique identifier may be used in the hash-to-UID table. Unneeded duplicate objects are deleted by copying their metadata to a manifest and then redirecting their unique identifiers to point at the manifest.

Growing community of inventors

Corpus Christi, TX, United States of America

Russell Turpin