Abstract: A system and method for managing data storage. A computer system comprises a data storage application and an API. The API presents a logical storage model to the data storage application for storing data on one or more storage appliances and accepts first and second plugins coupling first and second storage appliances, respectively, to the API. The first and second storage appliances store data according to different physical storage models. The API receives an access request from the data storage application targeting a portion of the logical storage model. In response to the access request, the API identifies a plugin and a corresponding storage appliance associated with the portion of the logical storage model and utilizes the identified plugin to map the portion of the logical storage model to a corresponding portion of the selected storage appliance's physical storage model.

Abstract: A system for efficient creation of aggregate backup images includes a plurality of backup images of a source data set, including a first and a second backup image, and a backup manager. The second backup image may include at least one data block containing an updated version of a corresponding data block in the first backup image. The backup manager may be configured maintain a first and a second image map corresponding to the first and second backup images, respectively, where each image map includes a plurality of entries pointing to data blocks of the corresponding backup image. In response to a request to create a first aggregate backup image of the source data set, the backup manager may be configured to establish a first aggregate image map including indirect pointers that point to one or more entries in each of the first and second image maps.

Abstract: A system for distributed discovery and management of frozen images includes a first and a second computer host, a first and a second frozen image agent and a frozen image server. Each of the first and second frozen image agents may be configured to identify one or more storage hierarchies hosted at a respective computer host, and to provide an encoding representing the identified storage hierarchies to the frozen image server. The frozen image server may be configured to aggregate the encodings provided by the frozen image agents into a system-level storage hierarchy catalog.

Abstract: A deleted snapshot of a data volume may be reconstructed from an existing snapshot and one or more modification logs. For example, modifications to the data volume that occur subsequent to creation of a first snapshot may be logged in a modification log. If the first snapshot is deleted, the first snapshot may be reconstructed by modifying a second snapshot, created after the first snapshot, according to the modification log in order to undo any modifications made to the data volume prior to creation of the second snapshot and after creation of the first snapshot.