Caringo, Inc.

Andrew David Klager

Russell Turpin

David Yoakley

Jonathan Ring

Don Baker

Aaron Pierce

Paul R M Carpentier

Paul Rm Carpentier

Donald George Baker

Tom Teugels

Patrick Murphy Ray

Thomas William Cook

Segments of a local range can appear multiple times in the sequence of erasure sets in a manifest without the data being duplicated. A subset of the content of an erasure set can be included in the logical object. It is not required that all erasure sets in an object use the same encoding. A manifest can indicate one or more portions of the logical object for which no data has been provided, and for which a 'fill' pattern is specified. A portion of data can be inserted at any point inside the object by inserting a reference to one or more segment sets containing that data into the manifest. If the desired insertion point is within an existing segment set, then that manifest set can be 'split' by incorporating two references to it, with the first specifying the range of data before the insertion point, and the second specifying the range of data after the insertion point. Many operations are thus enabled on an erasure-coded object in Object Storage such as 'append', &#x201c;insert,&#x201d; &#x201c;overwrite&#x201d; and &#x201c;merge.&#x201d;

Multi-part upload and editing of erasure-coded objects

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

Growing community of inventors

Austin, TX, United States of America

Andrew David Klager