Database failure recovery in an information handling system
An information handling system includes a plurality of backup devices, and a host device. The backup devices stores backup copies of data in the information handling system. The host device includes a database to store the data for the information handling system, and restore circuitry. The restore circuitry receives a restore event notification indicating a failure of the database. The restore circuitry then determines a number of backup copies of the data in the backup devices, and selects a combination of the backup devices, including first and second backup devices, to restore the data to the database based on the number of backup copies. The restore circuitry also restores the data to the database via a first thread associated with the first backup device, and a second thread associated with the second backup device, and the first and second threads are executed in parallel.
1. A method comprising:
receiving, at restore circuitry, a restore event notification that identifies a failure of a database of an information handling system;
determining, by the restore circuitry, a number of backup copies for the database;
deriving a restore time for each of a plurality of backup devices based on characteristics of the backup devices;
deriving a total restore time for each of a plurality of combinations of backup devices based on the derived total restore time for each of the backup devices;
selecting one combination of the backup devices from the combination of backup devices to restore data to the database during a restore operation based on the number of backup copies, wherein the combination of the backup devices includes first and second backup devices; and
restoring the data to the database via a first thread associated with the first backup device, and via a second thread associated with the second backup device, wherein the first and second threads are executed in parallel.
2. The method of claim 1 , further comprising:
detecting the failure of the database; and
providing the restore event notification in response to the failure of the database.
3. The method of claim 1 , further comprising:
determining characteristics of the backup devices associated with the number of backup copies; and
wherein the deriving the restore time for each of the backup devices based on the characteristics of the backup devices was performed prior to selecting the combination of the backup devices.
4. The method of claim 3 , further comprising:
determining an ordered list of the backup devices based on the derived total restore time for each of the backup devices.
5. The method of claim 1 , wherein the selected combination of the backup devices has a lowest total restore time of all of the combination of the backup devices.
6. The method of claim 3 , wherein the characteristics include latency of each of the backup devices, and throughput of each of the backup devices.
7. The method of claim 1 , further comprising:
providing a restore complete notification in response to the data being restored to the database.
8. An information handling system comprising:
a plurality of backup devices, the backup devices to store backup copies of data in the information handling system; and
a host device including:
a database to store the data for the information handling system; and
restore circuitry to communicate with the database and with the backup devices, the restore circuitry to receive a restore event notification indicating a failure of the database, to determine a number of backup copies of the data in the backup devices, to determine percentages of data to be retrieved from each backup device, to select a combination of the backup devices to restore the data to the database based on the number of backup copies and the percentages of data to be retrieved from each backup device, wherein the combination of the backup devices includes first and second backup devices, and to restore the data to the database via a first thread associated with the first backup device and via a second thread associated with the second backup device, wherein the first and second threads are executed in parallel.
9. The information handling system of claim 8 , wherein the host device further comprises:
redundancy circuitry to communication with the database, the redundancy circuitry to detect the failure of the database, and to provide the restore event notification in response to the failure of the database.
10. The information handling system of claim 8 , the restore circuitry further to determine characteristics of each of the backup devices associated with the number of backup copies, and to derive a restore time for each of the backup devices based on the characteristics of the backup devices prior to selecting the combination of the backup devices.
11. The information handling system of claim 10 , the restore circuitry further to derive a total restore time for each of a plurality of combinations of the backup devices based on the derived restore time for each of the backup devices.
12. The information handling system of claim 11 , wherein the selected combination of the backup devices has a lowest total restore time of all of the combination of the backup devices.
13. The information handling system of claim 10 , wherein the characteristics include latency of each of the backup devices, and throughput of each of the backup devices.
14. The information handling system of claim 8 , the restore circuitry further to provide a restore complete notification in response to the data being restored to the database.
15. An information handling system comprising:
a plurality of backup devices, the backup devices to store backup copies of data in the information handling system; and
a host device including:
a database to store the data for the information handling system
backup circuitry to communicate with the database and with the backup devices, the backup circuitry to transfer the data from the database to the backup devices, and to store the data as the backup copies in the backup devices; and
restore circuitry to communicate with the database and with the backup devices, the restore circuitry to receive a restore event notification indicating a failure of the database, to determine a priority level of failed logical units of the database, to determine a number of backup copies of the data in the backup devices, to select a combination of the backup devices to restore the data to the database based on the number of backup copies, to determine characteristics of each of the backup devices associated with the number of backup copies, to derive a restore time for each of the backup devices based on the characteristics of the backup devices prior to selecting the combination of the backup devices, to derive a total restore time for each of a plurality of combinations of the backup devices based on the derived restore time of each of the backup devices, wherein the restore begins with a highest priority level logical unit of the failed logical units, wherein the combination of the backup devices includes first and second backup devices, and to restore the data to the database via a first thread associated with the first backup device, and a second thread associated with the second backup device, wherein the first and second threads are executed in parallel.
16. The information handling system of claim 15 , wherein the selected combination of the backup devices has a lowest total restore time of all of the combination of the backup devices.
17. The information handling system of claim 15 , wherein the characteristics include latency of each of the backup devices, and throughput of each of the backup devices.
18. The information handling system of claim 15 , the restore circuitry further to provide a restore complete notification in response to the data being restored to the database.