Method, device and computer program product for job management

1. A method for job management, comprising:

sampling backup speeds for parallel executing group of historical concurrent backup jobs to obtain a trained goal ground truth backup speed;

obtaining an execution plan associated with a plurality of backup jobs including a target backup job having backup data, the execution plan at least indicating a size of backup data of the target backup job and start times of the plurality of backup jobs, submitted by users;

determining, based on the execution plan, a first set of backup jobs to be executed in parallel at a start time of the target backup job;

obtaining metrics of the first set of backup jobs, the metrics including variable and fixed blocks sizes;

determining, via function fitting, a predicted backup speed of executing the first set of backup jobs in parallel at the start time of the target backup job, based on the metrics and the ground truth backup speeds; and

determining, at least based on the predicted backup speed and the size of the backup data of the target backup job, time required for executing the target backup job.

2. The method of claim 1 , wherein determining the first set of backup jobs comprises:

determining, from the plurality of backup jobs, a preceding backup job with a start time prior to a start time of the target backup job;

determining, based on the start time and a size of backup data of the preceding backup job, whether the preceding backup job is to be completed before the start time of the target backup job; and

in response to determining that the preceding backup job fails to be completed before the start time of the target backup job, adding the preceding backup job into the first set of backup jobs.

3. The method of claim 1 , wherein determining the predicted backup speed comprises:

obtaining a metric associated with the first set of backup jobs, the metric comprising at least one of the following: a number of backup jobs using storage blocks of a fixed size in the first set of backup jobs, a number of backup jobs using storage blocks of variable sizes in the first set of backup jobs, a number of backup jobs with a specific backup cycle in the first set of backup jobs, and a number of concurrent threads of the first set of backup jobs; and

determining the predicted backup speed based on the metric.

4. The method of claim 3 , wherein determining the predicted backup speed based on the metric comprises:

processing the metric with a speed determination model to determine the predicted backup speed, wherein the speed determination model is a machine learning model trained based on metrics determined from a set of reference concurrent backup jobs and a corresponding ground-truth backup speed.

5. The method of claim 4 , further comprising: determining the ground-truth backup speed by sampling backup speeds of concurrently executing the set of reference concurrent backup jobs to obtain a set of sampled backup speeds;

determining at least one sampled backup speed based on a difference between a sampled backup speed and a further sampled backup speed at an adjacent time instant in the set of sampled backup speeds; and

determining the ground-truth backup speed based on the at least one sampled backup speed.

6. The method of claim 1 , wherein the predicted backup speed is a first predicted backup speed, and determining the time required for executing the target backup job comprises:

in response to completing at least one of the first set of backup jobs at a middle time or beginning to execute an additional backup job at the middle time, determining a second set of backup jobs to be executed in parallel at the middle time;

determining a second predicted backup speed of concurrently executing the second set of backup jobs; and

determining, based on the first predicted backup speed and the second predicted backup speed, the time required for executing the target backup job.

7. The method of claim 1 , further comprising:

providing time in form of at least one of: text, graphics, audio or video.

8. An electronic device, comprising:

at least one processing unit; and

at least one memory being 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:

sampling backup speeds for parallel executing group of historical concurrent backup jobs to obtain a trained goal ground truth backup speeds;

obtaining an execution plan associated with a plurality of backup jobs including a target backup job having backup data, the execution plan at least indicating a size of backup data of the target backup job and start times of the plurality of backup jobs, submitted by users;

determining, based on the execution plan, a first set of backup jobs to be executed in parallel at a start time of the target backup job;

obtaining metrics of the first set of backup jobs, the metrics including variable and fixed blocks sizes;

determining, via function fitting, a predicted backup speed of executing the first set of backup jobs in parallel at the start time of the target backup job, based on the metrics and the ground truth backup speeds; and

determining, at least based on the predicted backup speed and the size of the backup data of the target backup job, time required for executing the target backup job.

9. The device of claim 8 , wherein determining the first set of backup jobs comprises:

determining, from the plurality of backup jobs, a preceding backup job with a start time prior to a start time of the target backup job;

determining, based on the start time and a size of backup data of the preceding backup job, whether the preceding backup job is to be completed before the start time of the target backup job; and

in response to determining that the preceding backup job fails to be completed before the start time of the target backup job, adding the preceding backup job into the first set of backup jobs.

10. The device of claim 8 , wherein determining the predicted backup speed comprises:

obtaining a metric associated with the first set of backup jobs, the metric comprising at least one of the following: a number of backup jobs using storage blocks of a fixed size in the first set of backup jobs, a number of backup jobs using storage blocks of variable sizes in the first set of backup jobs, a number of backup jobs with a specific backup cycle in the first set of backup jobs, and a number of concurrent threads of the first set of backup jobs; and

determining the predicted backup speed based on the metric.

11. The device of claim 10 , wherein determining the predicted backup speed based on the metric comprises:

processing the metric with a speed determination model to determine the predicted backup speed, wherein the speed determination model is a machine learning model trained based on metrics determined from a set of reference concurrent backup jobs and a corresponding ground-truth backup speed.

12. The device of claim 11 , the acts further comprising: determining the ground-truth backup speed by

sampling backup speeds of concurrently executing the set of reference concurrent backup jobs to obtain a set of sampled backup speeds;

determining at least one sampled backup speed based on a difference between a sampled backup speed and a further sampled backup speed at an adjacent time instant in the set of sampled backup speeds; and

determining the ground-truth backup speed based on the at least one sampled backup speed.

13. The device of claim 8 , wherein the predicted backup speed is a first predicted backup speed, and determining the time required for executing the target backup job comprises:

in response to completing at least one of the first set of backup jobs at a middle time or beginning to execute an additional backup job at the middle time, determining a second set of backup jobs to be executed in parallel at the middle time;

determining a second predicted backup speed of concurrently executing the second set of backup jobs; and

determining, based on the first predicted backup speed and the second predicted backup speed, the time required for executing the target backup job.

14. The device of claim 8 , the acts further comprising:

providing time in a form of at least one of: text, graphics, audio or video.

15. A computer program product stored on a non-transitory computer storage medium and comprising machine-executable instructions which, when running in a device, cause the device to perform operations, the operations comprising:

sampling backup speeds for parallel executing group of historical concurrent backup jobs to obtain a trained goal ground truth backup speeds;

obtaining an execution plan associated with a plurality of backup jobs including a target backup job having backup data, the execution plan at least indicating a size of backup data of the target backup job and start times of the plurality of backup jobs, submitted by users;

determining, based on the execution plan, a first set of backup jobs to be executed in parallel at a start time of the target backup job;

obtaining metrics of the first set of backup jobs, the metrics including variable and fixed blocks sizes;

determining, via function fitting, a predicted backup speed of executing the first set of backup jobs in parallel at the start time of the target backup job, based on the metrics and the ground truth backup speeds; and

determining, at least based on the predicted backup speed and the size of the backup data of the target backup job, time required for executing the target backup job.

16. The computer program product of claim 15 , wherein determining the first set of backup jobs comprises:

determining, from the plurality of backup jobs, a preceding backup job with a start time prior to a start time of the target backup job;

determining, based on the start time and a size of backup data of the preceding backup job, whether the preceding backup job is to be completed before the start time of the target backup job; and

in response to determining that the preceding backup job fails to be completed before the start time of the target backup job, adding the preceding backup job into the first set of backup jobs.

17. The computer program product of claim 15 , wherein determining the predicted backup speed comprises:

obtaining a metric associated with the first set of backup jobs, the metric comprising at least one of the following: a number of backup jobs using storage blocks of a fixed size in the first set of backup jobs, a number of backup jobs using storage blocks of variable sizes in the first set of backup jobs, a number of backup jobs with a specific backup cycle in the first set of backup jobs, and a number of concurrent threads of the first set of backup jobs; and

determining the predicted backup speed based on the metric.

18. The computer program product of claim 17 , wherein determining the predicted backup speed based on the metric comprises:

processing the metric with a speed determination model to determine the predicted backup speed, wherein the speed determination model is a machine learning model trained based on metrics determined from a set of reference concurrent backup jobs and a corresponding ground-truth backup speed.

19. The computer program product of claim 18 , wherein the operations further comprise: determining the ground-truth backup speed by

sampling backup speeds of concurrently executing the set of reference concurrent backup jobs to obtain a set of sampled backup speeds;

determining at least one sampled backup speed based on a difference between a sampled backup speed and a further sampled backup speed at an adjacent time instant in the set of sampled backup speeds; and

determining the ground-truth backup speed based on the at least one sampled backup speed.

20. The computer program product of claim 15 , wherein the predicted backup speed is a first predicted backup speed, and determining the time required for executing the target backup job comprises:

in response to completing at least one of the first set of backup jobs at a middle time or beginning to execute an additional backup job at the middle time, determining a second set of backup jobs to be executed in parallel at the middle time;

determining a second predicted backup speed of concurrently executing the second set of backup jobs; and

determining, based on the first predicted backup speed and the second predicted backup speed, the time required for executing the target backup job.