Virtual machine replication strategy based on predicted application failures
Data protection operations including replication operations that dynamically adapt a topology of replica virtual machines. A data protection system may implement a machine model that is trained using, as input, characteristics of virtual machines. When a failure is predicted, a topology of the replica virtual machines is changed. The topology may also change when changes in the environment are detected. The changes may include redistributing the protected applications to the replica virtual machines and/or scaling the replica virtual machines.
1. A method for replicating data from a production site to a replica site, the method comprising:
repeatedly assessing, by a data protection system, applications protected by the data protection system based on a replication strategy using a machine model configured to generate a prediction regarding failure of the applications, wherein the replication strategy is configured to predict the failure based on inputs that include characteristics of the virtual machines;
monitoring for changes in the applications at the production site, wherein the changes include quality of service changes, failure domain changes, and/or related application changes;
generating a first topology of the replica virtual machines based on the prediction of the machine model;
replicating the applications from production virtual machines at the production site to the replica virtual machines according to the replication strategy, which includes the first topology; and
in response to a change in the prediction of the machine model or the change in the application, changing the first topology of the replica virtual machines to a second topology of the replica virtual machines based on the changed prediction of the machine model.
2. The method of claim 1 , further comprising performing sampling to collect sample data corresponding to the characteristics of the virtual machines.
3. The method of claim 2 , wherein the characteristics of the virtual machines include the one or more of disk IO latency, disk change rate, network IO latency, network change rate, processor utilization, memory utilization, IO error number, and/or IO error rate, wherein the machine model generates a value, wherein the topology is changed when the value exceeds a threshold value.
4. The method of claim 3 , wherein the sample data includes data points and time series data.
5. The method of claim 3 , further comprising training the machine model with the sample data.
6. The method of claim 3 , wherein the inputs further includes triggers regarding an incoming danger.
7. The method of claim 1 , further comprising adapting to the changes in the production site.
8. The method of claim 1 , further comprising changing the topology by moving applications, moving portable applications, and/or scaling the replica virtual machines.
9. The method of claim 8 , further comprising scaling the replica virtual machines by cloning a replica virtual machine, wherein only an operating disk is cloned to generate a new replica virtual machine and wherein applications are then moved to the new replica virtual machine.
10. The method of claim 8 , further comprising creating a replica virtual machine template, wherein scaling the replica virtual machines includes instantiating a new replica virtual machine based on the replica virtual machine template.
11. A non-transitory storage medium having stored therein instructions that are executable by one or more hardware processors to perform operations for replicating a production site to a replica site, the method comprising:
repeatedly assessing, by a data protection system, applications protected by the data protection system based on a replication strategy using a machine model configured to generate a prediction regarding failure of the applications, wherein the replication strategy is configured to predict the failure based on inputs that include characteristics of the virtual machines;
monitoring for changes in the applications at the production site, wherein the changes include quality of service changes, failure domain changes, and/or related application changes;
generating a first topology of the replica virtual machines based on the prediction of the machine model;
replicating the applications from production virtual machines at the production site to the replica virtual machines according to the replication strategy, which includes the first topology; and
in response to a change in the prediction of the machine model or the change in the application, changing the first topology of the replica virtual machines to a second topology of the replica virtual machines based on the changed prediction of the machine model.
12. The non-transitory storage medium of claim 11 , the operations further comprising performing sampling to collect sample data corresponding to the characteristics of the virtual machines.
13. The non-transitory storage medium of claim 12 , wherein the characteristics of the virtual machines include the one or more of disk IO latency, disk change rate, network IO latency, network change rate, processor utilization, memory utilization, IO error number, and/or IO error rate, wherein the machine model generates a value, wherein the topology is changed when the value exceeds a threshold value.
14. The non-transitory storage medium of claim 13 , wherein the sample data includes data points and time series data.
15. The non-transitory storage medium of claim 13 , the operations further comprising training the machine model with the sample data.
16. The non-transitory storage medium of claim 13 , wherein the inputs further includes triggers regarding an incoming danger.
17. The non-transitory storage medium of claim 11 , the operations further comprising adapting to the changes in the production site.
18. The non-transitory storage medium of claim 11 , the operations further comprising changing the topology by moving applications, moving portable applications, and/or scaling the replica virtual machines.
19. The non-transitory storage medium of claim 18 , the operations further comprising scaling the replica virtual machines by cloning a replica virtual machine, wherein only an operating disk is cloned to generate a new replica virtual machine and wherein applications are then moved to the new replica virtual machine.
20. The non-transitory storage medium of claim 18 , the operations further comprising creating a replica virtual machine template, wherein scaling the replica virtual machines includes instantiating a new replica virtual machine based on the replica virtual machine template.