Distributed computing system with resource managed database cloning

Patent metadata

Field	Value
Publication number	US-9460181-B2
Application number	US-201614993726-A
Country	US
Kind code	B2
Filing date	Jan 12, 2016
Priority date	Jan 23, 2015
Publication date	Oct 4, 2016
Grant date	Oct 4, 2016

Abstract

In a distributed computing system, cloning operations determine when each of multiple backup database instances most recently experienced an update. If a most recently updated backup database instance was updated within a prescribed time period of one or more other backup database instances, a source instance for cloning is deemed to be one of these database instances satisfying a prescribed proximity criteria relative to a designated database instance. If a difference in update times is greater than the prescribed time period for the two most recent backup database instances, the source instance for cloning is deemed to be the most recently updated backup database instance. The control center conducts cloning to a target instance using the selected backup database instance as a source instance.

First claim

What is claimed is:

1. A computer-driven distributed data storage and management system for cloning database instances, comprising:
multiple datacenters comprising first and second datacenters;
wherein:
each of the datacenters comprises a plurality of computerized servers and digital data storage;
each of the servers comprises a digital data processor coupled to the digital data storage;
the digital data storage of the first datacenter comprises a primary database;
the digital data storage of the second datacenter comprises a standby database;

the system further comprising:
backup database instances constructed by copying data from a corresponding data source comprising the primary or the standby database; and

a control center coupled to the datacenters comprising a processor, a memory, a store and retrieve module, a cloning module, and other modules, wherein the modules comprise instructions stored in the control center memory that execute on the control center processor to execute operations;
wherein:
the store and retrieve module directs the datacenters, including operation of the primary database to store and retrieve data on behalf of remote clients, and operation of the standby database to mirror the primary database for use upon a failover event;
the cloning module, responsive to receiving or detecting a prescribed command, event trigger, or other occurrence:
identifies a plurality of the backup database instances and retrieves machine-readable records listing update times when each of the identified backup database instances most recently experienced an update from the backup database instance's data source;
responsive to a first identified backup database instance having an update time within a prescribed period of a second identified backup database instance, where the first identified backup database instance is a most recently updated of the identified backup database instances, selects, as a source instance for cloning, the first or the second identified backup database instances satisfying a prescribed proximity criteria relative to a designated database instance;
responsive to a difference in update times being greater than the prescribed period for two identified backup databases experiencing most recent updates, selects, as a source instance for cloning, one of the identified backup database instances whose update time is most recent; and
clones the selected source instance to a target instance.

2. The system of claim 1 , wherein the control center receives and implements user configuration of the prescribed period.

3. The system of claim 1 , wherein the control center receives and implements user configuration of the prescribed proximity criteria.

4. The system of claim 3 , wherein the proximity criteria comprises a physical proximity.

5. The system of claim 3 , wherein the proximity criteria comprises a logical proximity.

6. The system of claim 1 , wherein the prescribed proximity criteria favors backup database instances sharing a common datacenter as the designated data instance.

7. The system of claim 1 , wherein:
prior to the cloning operation, the control center conducts preparatory operations comprising conducting prescribed preflight checks, provisioning the target instance, and restoring from the selected backup instance to the provisioned target instance; and
responsive to failure of any of the preparatory operations, instead of cloning the selected backup database instance, the control center performs the cloning operation using only processing nodes associated with the primary or secondary database to copy data directly from the primary or secondary database.

8. The system of claim 1 , wherein:
prior to the cloning operation, the control center conducts preparatory operations comprising conducting prescribed preflight checks, provisioning the target instance, and restoring from the selected backup instance to the provisioned target instance; and
responsive to success of the preparatory operations, the control center conducts the cloning using only processing nodes unassociated with the primary and secondary databases.

9. The system of claim 7 , wherein the prescribed preflight checks include verifying that the selected backup database instance exists.

10. The system of claim 7 , wherein the preflight checks comprise validating the selected backup database and target instance in a configuration management database.

11. The system of claim 7 , wherein the preflight checks comprise verifying that the selected backup database instance is capable of being successfully restored.

12. The system of claim 1 , wherein the designated database instance comprises a previously created backup database instance to be replaced by the target instance.

13. The system of claim 1 , wherein the designated database instance comprises a testing instance, and the testing instance is replaced with the target instance.

14. The system of claim 1 , wherein the cloning comprises:
provisioning the target instance;
copying data from the selected backup instance to the provisioned target instance; and
re-pointing application nodes linked to a previously created backup database so that the application nodes link to the provisioned target instance.

15. The system of claim 14 , wherein the control center is programmed to permit roll back of the target instance to the previously created database for a given time period after completing the cloning.

16. The system of claim 1 , wherein the control center is programmed to perform a multiplicity of cloning operations to create a multiplicity of target instances.

17. A computer-implemented method of conducting a database clone in a distributed computing system comprising a control center comprising a processor, coupled to multiple datacenters, each of the datacenters comprising a plurality of computerized servers, each of the computerized servers comprising a digital data processing machine coupled to digital data storage, wherein digital data storage of a first datacenter comprises a primary database operated on behalf of remote clients, and wherein digital data storage of a second datacenter comprises a standby database that mirrors the primary database for use upon a failover event, wherein the digital data storage also comprises a backup database constructed by copying data from the primary or standby databases, wherein the method comprises computer-implemented operations of:
receiving or detecting, by the control center, a prescribed command, event trigger, or other occurrence;
responsive to the receiving or detecting, identifying, utilizing a processor of the control center, a plurality of the backup database instances and retrieving machine-readable records listing update times when each of the identified backup database instances most recently experienced an update from the backup database instance's data source;
responsive to a first identified backup database instance having an update time within a prescribed period of a second identified backup database instance, where the first identified backup database instance is a most recently updated of the identified backup database instances, the control center selecting, utilizing its processor, a source instance for cloning one of the first and second identified backup database instances satisfying a prescribed proximity criteria relative to a designated database instance;
responsive to a difference in update times being greater than the prescribed period for two identified backup databases experiencing most recent updates, the control center, utilizing

Assignees

• Servicenow Inc

Inventors

• Wang Paul
• Ye Xiaoyi
• Lu Xuejia
• Chandrashekar Sridhar

Classifications

• G06F11/1461

  Backup scheduling policy · CPC title

• G06F11/1469

  Backup restoration techniques · CPC title

• G06F2201/80

  Database-specific techniques · CPC title

• G06F11/1464 Primary

  for networked environments · CPC title

• G06F16/27

  Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor · CPC title