Dell PowerStore 3000X EMC PowerStore Protecting Your Data - Page 11

Failover You can fail over a replication session to change the role of the destination system to the source system. After a failover, you can access applications on the new source system to recover data. There are two types of failovers: planned and unplanned failovers. Planned failover In a planned failover, you manually fail over a replication session from the source system to the destination system. The destination system is synchronized with the source system before the failover, and there is no data loss. Before you perform a planned failover, ensure that you stop I/O operations for any applications and hosts. You cannot pause a replication session that is undergoing a planned failover. During a planned failover, you can take the following actions: • Perform an unplanned failover • Delete the replication session by removing the protection policy on the storage resource After a planned failover, the replication session is inactive. Reprotect the replication session to synchronize the destination storage resource, and resume the replication session. You can also use the auto-reprotect feature before failing over, which reverses the direction of replication between the source and the target after the failover is complete. Unplanned failover Unplanned failovers occur after events such as source system failure, or an event on the source system that leads to downtime for production access. You start the unplanned failover from the destination system, and provide production access to the original destination resource from a synchronized point-in-time snapshot. When the connection to the source system is reestablished, the original source resource is placed into destination mode. After an unplanned failover, you can restore the system from the latest data or any point-in-time snapshot after a restore. Reprotect the replication session to synchronize the destination storage resource, and then resume the replication session. Recovery point objective Recovery point objective (RPO) indicates the acceptable amount of data, measured in units of time, that may be lost in a failure. When you set up a replication rule, you can configure automatic synchronization based on the RPO. You can specify an RPO from a minimum of 5 minutes up to a maximum of 24 hours. The default RPO is set at one hour. NOTE: Although a smaller RPO interval provides more protection and consumes less space, it has a higher performance impact, resulting in more network traffic. A higher RPO interval may result in more space consumption, which can affect snapshot schedules and space thresholds. You can also specify an alert threshold, which is an acceptable delay added to the RPO to determine if a replication compliance alert should be generated. The system generates an alert if the delay between the RPO and actual synchronization time exceeds the alert threshold. Setting the alert threshold to zero means that alerts will be generated if the actual synchronization time exceeds the RPO. Synchronization Synchronization asynchronously updates the destination resource with changes on the source resource from the previous synchronization cycle. Size changes, membership changes, or both, are also synchronized on the source resource. Synchronization happens either automatically - according to a set schedule - or manually. Snapshots are synchronized from the source system to the destination system, and maintain block sharing efficiency. You can synchronize a replication session when it is in the following states: • Operating normally • System paused While a replication session is synchronizing, you can take the following actions: • Planned failover from the source system • Fail over from the destination system • Pause replication sessions from the source or destination system • Delete a replication session by removing a protection policy Replication 11