Method, apparatus and computer program product for managing backup system

1. A method of managing a backup system, comprising:

obtaining a state of a backup system, the backup system comprising a backup server and at least one backup client, the backup server being communicatively coupled to the at least one backup client via a network and configured to back up data of the at least one backup client;

determining a reward score corresponding to the state of the backup system, wherein the backup server comprises a plurality of backup tasks for performing data backups on the at least one backup client, and the state of the backup system indicates a duration and result of a last execution of a given backup task of the plurality of backup tasks, and determining the reward score comprises: determining the reward score based on the duration and result of the last execution of the given backup task, wherein the reward score is inversely proportional to a completion time of the given backup task in the backup system;

determining, based on the state of the backup system and the reward score, configuration information for the backup system; and

reconfiguring, based on the configuration information for the backup system, a backup schedule for the backup server to perform data backups on the at least one backup client in real time.

2. The method of claim 1 , wherein the state of the backup system comprises at least one of the following:

a server state of the backup server;

respective client states of the at least one backup client; and

a network state of the network.

3. The method of claim 2 , wherein the server state comprises at least one of the following:

storage usage at the backup server;

CPU usage at the backup server;

memory usage at the backup server; and

input/output operation usage at the backup server.

4. The method of claim 2 , wherein a client state of a backup client from the at least one backup client comprises at least one of the following:

CPU usage at the backup client;

memory usage at the backup client;

input/output operation usage at the backup client;

the number of files at the backup client;

a size of data at the backup client;

a duration of a last data backup performed on the backup client; and

a result of the last data backup performed on the backup client.

5. The method of claim 2 , wherein the network state comprises at least one of the following:

round-trip delay of the network; and

an available network bandwidth of the network.

6. The method of claim 1 , wherein determining the configuration information comprises:

determining, based on the state of the backup system and the reward score, the configuration information using a neural network.

7. The method of claim 6 , wherein the neural network comprises an actor network and a critic network, the actor network comprises an actor evaluation network and an actor target network, the critic network comprises a critic evaluation network and a critic target network, and the neural network further comprises a first sub-network for updating a first network parameter of the actor evaluation network and a second sub-network for updating a second network parameter of the critic evaluation network, and wherein:

the actor evaluation network is configured to determine, based on the state of the backup system obtained at a current time and the first network parameter of the actor evaluation network, the configuration information for the backup system at the current time, the configuration information being provided to the critic evaluation network;

the actor target network is configured to determine, based on the state of the backup system obtained at the current time, target configuration information for the backup system at the current time, the target configuration information being provided to the critic target network;

the critic evaluation network is configured to determine, based on the state of the backup system obtained at the current time and the second network parameter of the critic evaluation network, an evaluation of the configuration information, the evaluation being provided to the first sub-network and the second sub-network;

the critic target network is configured to determine, based on the state of the backup system obtained at the current time, a target evaluation of the target configuration information, the target evaluation being provided to the second sub-network;

the first sub-network is configured to update the first network parameter based on the evaluation; and

the second sub-network is configured to update the second network parameter based on the reward score and a difference between the evaluation and the target evaluation.

8. The method of claim 6 , wherein the neural network is based on a Deep Deterministic Policy Gradient (DDPG) algorithm for implementing a continuous control, the DDPG algorithm being designed based on a Deep Q Network (DQN).

9. The method of claim 1 , wherein the configuration information is represented by a 1×N array, N representing a number of backup clients including the at least one backup client, each element in the 1×N array indicating an execution time of a corresponding backup task.

10. An electronic device comprising:

at least one processing unit;

at least one memory coupled to the at least one processing unit and storing instructions for execution by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the device to perform acts comprising:

obtaining a state of a backup system, the backup system comprising a backup server and at least one backup client, the backup server being communicatively coupled to the at least one backup client via a network and configured to back up data of the at least one backup client;

determining a reward score corresponding to the state of the backup system, wherein the backup server comprises a plurality of backup tasks for performing data backups on the at least one backup client, and the state of the backup system indicates a duration and result of a last execution of a given backup task of the plurality of backup tasks, and determining the reward score comprises: determining the reward score based on the duration and result of the last execution of the given backup task, wherein the reward score is inversely proportional to a completion time of the given backup task in the backup system;

determining, based on the state of the backup system and the reward score, configuration information for the backup system; and

reconfiguring, based on the configuration information for the backup system, a backup schedule for the backup server to perform data backups on the at least one backup client in real time.

11. The device of claim 10 , wherein the state of the backup system comprises at least one of the following:

a server state of the backup server;

respective client states of the at least one backup client; and

a network state of the network.

12. The device of claim 11 , wherein the server state comprises at least one of the following:

storage usage at the backup server;

CPU usage at the backup server;

memory usage at the backup server; and

input/output operation usage at the backup server.

13. The device of claim 11 , wherein a client state of a backup client from the at least one backup client comprises at least one of the following:

CPU usage at the backup client;

memory usage at the backup client;

input/output operation usage at the backup client;

the number of files at the backup client;

a size of data at the backup client;

a duration of a last data backup performed on the backup client; and

a result of the last data backup performed on the backup client.

14. The device of claim 11 , wherein the network state comprises at least one of the following:

round-trip delay of the network; and

an available network bandwidth of the network.

15. The device of claim 10 , wherein determining the configuration information comprises:

determining, based on the state of the backup system and the reward score, the configuration information using a neural network.

16. The device of claim 15 , wherein the neural network comprises an actor network and a critic network, the actor network comprises an actor evaluation network and an actor target network, the critic network comprises a critic evaluation network and a critic target network, and the neural network further comprises a first sub-network for updating a first network parameter of the actor evaluation network and a second sub-network for updating a second network parameter of the critic evaluation network, and wherein:

the actor evaluation network is configured to determine, based on the state of the backup system obtained at a current time and the first network parameter of the actor evaluation network, the configuration information for the backup system at the current time, the configuration information being provided to the critic evaluation network;

the actor target network is configured to determine, based on the state of the backup system obtained at the current time, target configuration information for the backup system at the current time, the target configuration information being provided to the critic target network;

the critic evaluation network is configured to determine, based on the state of the backup system obtained at the current time and the second network parameter of the critic evaluation network, an evaluation of the configuration information, the evaluation being provided to the first sub-network and the second sub-network;

the critic target network is configured to determine, based on the state of the backup system obtained at the current time, a target evaluation of the target configuration information, the target evaluation being provided to the second sub-network;

the first sub-network is configured to update the first network parameter based on the evaluation; and

the second sub-network is configured to update the second network parameter based on the reward score and a difference between the evaluation and the target evaluation.

17. A computer program product tangibly stored on a non-transitory computer storage medium and including machine-executable instructions which cause, when executed by a device, the device to perform operations, the operations comprising:

obtaining a state of a backup system, the backup system comprising a backup server and at least one backup client, the backup server being communicatively coupled to the at least one backup client via a network and configured to back up data of the at least one backup client;

determining a reward score corresponding to the state of the backup system, wherein the backup server comprises a plurality of backup tasks for performing data backups on the at least one backup client, and the state of the backup system indicates a duration and result of a last execution of a given backup task of the plurality of backup tasks, and determining the reward score comprises: determining the reward score based on the duration and result of the last execution of the given backup task, wherein the reward score is inversely proportional to a completion time of the given backup task in the backup system;

determining, based on the state of the backup system and the reward score, configuration information for the backup system; and

reconfiguring, based on the configuration information for the backup system, a backup schedule for the backup server to perform data backups on the at least one backup client in real time.

18. The computer program product of claim 17 , wherein the state of the backup system comprises at least one of the following:

a server state of the backup server;

respective client states of the at least one backup client; and

a network state of the network.

19. The computer program product of claim 18 , wherein the server state comprises at least one of the following:

storage usage at the backup server;

CPU usage at the backup server;

memory usage at the backup server; and

input/output operation usage at the backup server.

20. The computer program product of claim 18 , wherein a client state of a backup client from the at least one backup client comprises at least one of the following:

CPU usage at the backup client;

memory usage at the backup client;

input/output operation usage at the backup client;

the number of files at the backup client;

a size of data at the backup client;

a duration of a last data backup performed on the backup client; and

a result of the last data backup performed on the backup client.