DATA PROTECTION SCHEDULING, SUCH AS PROVIDING A FLEXIBLE BACKUP WINDOW IN A DATA PROTECTION SYSTEM

• US 20140279922A1
• Filed: 06/03/2014
• Published: 09/18/2014
• Est. Priority Date: 06/18/2008
• Status: Active Grant

• Current Assignee: CommVault Systems Incorporated
• Original Assignee: CommVault Systems Incorporated
• Inventors: Kottomtharayil, Rajiv
• Granted Patent: US 10,198,324 B2
• Time in Patent Office: Days
• US Class Current: 707/654
• CPC Class Codes: G06F 11/1461 Backup scheduling policy
• CPC Class Codes: G06F 11/1464 for networked environments
• CPC Class Codes: G06F 2209/5019 Workload prediction
• CPC Class Codes: G06F 2209/506 Constraint
• CPC Class Codes: G06F 9/505 considering the load

Abstract

A data protection scheduling system provides a flexible or rolling data protection window that analyzes various criteria to determine an optimal or near optimal time for performing data protection or secondary copy operations. While prior systems may have scheduled backups at an exact time (e.g., 2:00 a.m.), the system described herein dynamically determines when to perform the backups and other data protection storage operations, such as based on network load, CPU load, expected duration of the storage operation, rate of change of user activities, frequency of use of affected computer systems, trends, and so on.

Claims

1. A system for dynamically scheduling storage operations so that the storage operations can be completed during an available storage window, the system comprising:

a management component to receive storage operation requests;

a jobs agent configured to dynamically determine a schedule for performing multiple storage operations;

a monitoring component configured to monitor usage of resources after starting storage operations, and to assist in dynamically altering the schedule for performing storage operations; and

a database agent configured to store performance data for determining a time to perform each storage operation,

wherein the performance data facilitates an analysis of system resource load to be used at least in part in determining the schedule; and

an analysis component configured to determine when storage operations are exempt from dynamically altering scheduling,

wherein when a storage operation is exempt from dynamically altered scheduling, then the system is configured to—

perform the storage operation in an order in which hard coded storage operations to be completed within a time window are performed first,

followed by performing storage operations having a deadline of completion less than a threshold deadline.

2. The system of claim 1 wherein the jobs agent determines a priority for a received storage operation request based on the stored performance data.

3. The system of claim 1 wherein the jobs agent dynamically determines a schedule based on the resources required by each request and usage of the resources at various times within the storage window.

4. The system of claim 1 wherein the database agent stores historical data about an availability of resources for performing storage operations.

5. The system of claim 1 wherein the database agent stores recently measured data about an availability of resources for performing storage operations.

6. The system of claim 1 wherein the jobs agent dynamically determines a schedule that increases the priority of previously scheduled jobs that were not completed in a previous storage window.

7. A method in a data storage system for performing a data storage operation, the method comprising:

within a selected time window between a start time and an end time, determining that a condition associated with a data storage operation to be performed by the data storage system satisfies a threshold condition;

if the threshold condition is satisfied, then scheduling a time to start the data storage operation within the selected time window, based on the determined condition;

performing at least a portion of the data storage operation at the scheduled time within the selected time window; and

monitoring usage of resources after starting the data storage operation to determine whether to dynamically alter the schedule for performing the storage operation,

wherein one or more applied performance criteria relate to a predicted usage of computing resources in performing the storage operation;

wherein the scheduling further comprises;

determining the selected time window is selected at least partly based on a future network load;

determining a schedule that is sufficient to complete multiple requested data storage operations within the storage window,

comparing, during execution of the schedule, the storage window with an estimated time remaining to complete data storage operations;

delaying, based on the comparison, storage of some types of data in order to store other types of data within the storage window; and

sending some data storage jobs to other data storage systems.

8. The method of claim 7, wherein the condition associated with the data storage operation relates to information included in a request to perform the data storage operation.

9. The method of claim 7, wherein the condition associated with the data storage operation relates to characteristics of resources within the data storage system.

10. The method of claim 7, wherein the condition is calculated from multiple criteria, the criteria including:

a first criterion associated with a relative priority to the data storage operation; and

a second criterion associated with a determined usage of resources within the data storage system.

11. The method of claim 7, wherein the threshold condition is dynamically determined based on a characteristics of the performing data storage operation.

12. The method of claim 7, further comprising:

stopping the data storage operation at a time after the scheduled time when the condition does not satisfy the threshold condition; and

restarting the data storage operation at a next time after the scheduled time when the condition satisfies the threshold condition.

13. A system for instructing a data storage system to perform data storage operations outside of a scheduled time window, comprising:

a resource monitoring component, wherein the resource monitoring component receives information identifying usage of resources within a data storage system outside of the time window,

a priority component, wherein the priority component assigns a relative priority to two or more jobs within a queue of jobs to be transferred to storage media;

a determination component, wherein the determination component determines when a threshold condition associated with the usage of resources within the data storage system is satisfied; and

a scheduling component, wherein the scheduling component schedules the jobs to be performed by the data storage system based on the assigned relative priorities when the threshold condition is satisfied.

14. At least one tangible computer memory storing a data structure, wherein the data structure includes two or more entries, each of the entries relating to a data storage job to be performed by a data storage system and comprising:

a first criterion associated with one or more characteristics of the data storage job to be performed by a data storage system;

a second criterion associated with one or more characteristics of resources within the data storage system; and

a third criterion associated with historical information related to a previous performance of the resources within the data storage system;

wherein the data structure is configured to be accessed by a storage policy within the data storage system before performance of the data storage job.

15. The computer memory of claim 14, wherein the first criterion includes:

an assigned priority for the data storage job relative to other data storage jobs within the data structure;

a deadline in which to perform the data storage job; and

a number of times the data storage job was previously scheduled to be performed.

16. The computer memory of claim 14, wherein the second criterion includes:

a value for a current usage of the resources within the data storage system; and

a value for a predicted usage of the resources within the data storage system related to performing the data storage job using the resources.