Parallel database backup and restore

A method, implemented at a computer system that includes one or more processors, for simultaneously restoring data from a single backup node to a plurality of data storage nodes, the method comprising the following:

• an act of the computer system receiving a restore command from a user, the restore command indicating that a specified portion of data of a parallel database is to be restored from an existing data backup at a backup node, wherein the data backup is stored in a data store at the backup node;
• based on receiving the restore command, an act of the computer system generating a plurality of restore commands, including generating a different restore command for each of a plurality of data storage nodes of the parallel database, each different restore command being configured to instruct the corresponding data storage node to initiate a restore of a corresponding piece of the specified portion of data that is stored at the existing data backup, such that the plurality of restore commands are configured to initiate a parallel restore from the existing data backup in the data store to each of the plurality of data storage nodes of the parallel database;
• and
• an act of the computer system initiating execution of the generated plurality of restore commands at each of the plurality of data storage nodes in a parallel manner, such that each corresponding piece of data from the existing data backup in the data store is restored in parallel to each corresponding data storage node.

A computer program product comprising one or more computer-readable hardware storage devices having stored thereon computer-executable instructions that, when executed by one or more processors of a computer system, cause the computer system to implement a method for simultaneously restoring data from a single backup node to a plurality of data storage nodes, the method comprising the following:

• receiving a restore command from a user, the restore command indicating that a specified portion of data of a parallel database is to be restored from an existing data backup at a backup node, wherein the data backup is stored in a data store at the backup node;
• based on receiving the restore command, generating a plurality of restore commands, including generating a different restore command for each of a plurality of data storage nodes of the parallel database, each different restore command being configured to instruct the corresponding data storage node to initiate a restore of a corresponding piece of the specified portion of data that is stored at the existing data backup, such that the plurality of restore commands are configured to initiate a parallel restore from the existing data backup in the data store to each of the plurality of data storage nodes of the parallel database;
• and
• initiating execution of the generated plurality of restore commands at each of the plurality of data storage nodes in a parallel manner, such that each corresponding piece of data from the existing data backup in the data store is restored in parallel to each corresponding data storage node.

A computer system comprising the following:

• one or more processors;
• system memory;
• and
• one or more computer-readable storage media having stored thereon computer-executable instructions that, when executed by the one or more processors, cause the computing system to simultaneously back up data from a plurality of data storage nodes to a backup node, including at least the following:
• receiving a backup command indicating that a specified portion of data of a parallel database is to be backed up to a backup node, wherein the specified portion of data is stored across a plurality of different data storage nodes of the parallel database;
• based on receiving the backup command, generating a plurality of backup commands, including generating a different backup command for each of the plurality of data storage nodes, each different backup command being configured to instruct the corresponding data storage node to initiate a backup of a corresponding piece of the specified portion of data that is stored at the corresponding data storage node, such that the plurality of backup commands are configured to initiate a parallel backup from the plurality of data storage nodes to a single data store at the backup node;
• and
• initiating execution of the generated plurality of backup commands at each of the plurality of data storage nodes in a parallel manner, such that the corresponding piece of data from each of the plurality of different data storage nodes is backed up in parallel to the data store at the backup node, wherein a portion of corresponding metadata is stored with the backup data to indicate which data storage node each portion of data was received from.