Abstract: A backup of an application stored in a container on a backup media server is accessed. The container is formatted as a virtual hard disk and in a read-only mode. A differencing virtual hard disk is created at a recovery server. The differencing disk is pointed to the container. A size of the differencing disk is increased. The differencing disk is mounted onto the recovery server. Writes are allowed to be made to the differencing disk mounted at the recovery server while the virtual hard disk formatted container remains in the read-only mode on the backup media server.

Abstract: In one example, a method for data backup includes surveying a database availability group (DAG) that includes multiple nodes which are each associated with one or more databases, to determine a distribution of the databases across the nodes. Next, federated logic is used to create a profile of the DAG based on the survey, and the profile is used to generate a dynamic preferred server order list (PSOL). Load balancing for the DAG is performed using the dynamic PSOL, and the dynamic PSOL is updated based on results of the load balancing. Finally, a federated backup of the databases is performed based on the updated dynamic PSOL.

Abstract: Embodiments described herein use a file server implementing a network file sharing protocol in an improved backup workflow in a storage system. The file server provides storage services to files stored in a plurality of storage devices via one or more file sharing protocols. In one embodiment, in response to a request to back up data that is received at the file server from an application server, the file server creates a snapshot of the data to be backed up using data stored in a source storage device. The file server sends the created snapshot to the target storage system for storage and sends a first notification to the application server indicating that the snapshot has been backed up in the target storage system. In addition, the application server sends a second notification to a backup server indicating that the data has been backed up.

Abstract: In one example, a method for data backup includes surveying an environment, that has a plurality of nodes which are each associated with one or more databases, to determine a distribution of the databases across the nodes. A database profile is then created based on the survey and identifies a node with maximum ownership of databases. Next, a preferred server order list (PSOL) is generated that is based on the database profile, a last activation preference for one of the databases, and a failover preference for one of the databases. Using the PSOL, a backup workload is balanced across the nodes, and the databases are then backed up using the results of the backup workload balancing.