@@ -8,11 +8,9 @@ store metadata: [https://docs.openstack.org/mitaka/user-guide/cli\_config\_drive
...
@@ -8,11 +8,9 @@ store metadata: [https://docs.openstack.org/mitaka/user-guide/cli\_config\_drive
## Replication
## Replication
Replication is the act of continuous copying of the instance content \(more specifically \) to a separate storage. In case of hardware failure of the underlying infrastructure it is possible to spin-up your VM from this copy. This copy is not entirely identical to the original instance Although the copy is recorded periodically, there usually is a few minutes delay compared to the state of the live instance. This is an asynchronnous process.
Replication is the act of continuous copying of the instance content \(more specifically \) to a separate storage. In case of hardware failure of the underlying infrastructure it is possible to spin-up your VM from this copy. This copy is not entirely identical to the original instance Although the copy is recorded periodically, there usually is a few minutes delay compared to the state of the live instance. This is an asynchronnous process.
**High Availability**– the live instance is monitored non-stp and in case of physical hardware failure is immediately spinned up on another physical server.
**High Availability**– the live instance is monitored non-stp and in case of physical hardware failure is immediately spinned up on another physical server.
**Fault Tolerance**– This feature enables to run one virtual machine on multiuple physical servers. This feature is functionally close to replication and differs in the fact that it is implemented as a synchronnous process. This means that both replicas are identical all of the time. This feature brings about certain technical limitations. Please contact support in case you are interested in this feature, as some things need to be specified and negotiated.
**Fault Tolerance**– This feature enables to run one virtual machine on multiuple physical servers. This feature is functionally close to replication and differs in the fact that it is implemented as a synchronnous process. This means that both replicas are identical all of the time. This feature brings about certain technical limitations. Please contact support in case you are interested in this feature, as some things need to be specified and negotiated.