Before you can understand the details of the integration between Arcserve Backup and Arcserve Replication, you must be familiar with some of the terms and definitions used by each product.
The integration uses the following terms and definitions:
An Arcserve Replication scenario is the fundamental basis for managing the operation of the system. An Arcserve Replication scenario always includes at least one Master server and one Replica server. In addition, multiple independent scenarios can run on a single server.
A scenario is a structure that describes the following:
Note: Arcserve Backup supports File Server, MS Exchange, and SQL Server scenarios only.
The Master server is the active or production server that lets you actively change (read and write) data. Any changes made at any given time on the Master server are captured continuously in real time and transferred (or replicated) to one or more of the associated Replica servers so that all the Replica servers contain an exact copy of the data on the Master server at all times.
The Replica server is the passive server. This is the server from which data cannot be changed (read only) in any way except through changes replicated from the Master server. There can be multiple Replica servers associated with a single Master server. When synchronizing the replicated data with the Master server, the data between the servers is compared and only the changes made to the Master server are sent to the Replica servers, which minimizes WAN traffic.
Continuous data protection (CDP) is the ability to recover data not just to certain isolated previous states captured, for example, in a daily or weekly backup or snapshot, but to recover the data back to any point in time. That way, whenever a virus occurs, you can recover to a point just minutes before the virus occurred with essentially zero data loss and a very fast recovery time.
Failover is a feature that detects when the protected application on the Master server fails and switches (either automatically or manually activated) to a designated Replica server with essentially zero loss of data and time. If a Master server fails or must be shut down for maintenance, a synchronized replica (locally or in a remote location) will instantly and automatically take its place.
Data rewind is a recovery method that allows rewinding files to a point in time (rewind point) before they were corrupted. This built-in rewind technology occurs on the Replica server and any "data rewinding" to a previous state can be performed on the Replica server only.
A rewind point is a checkpoint in the Rewind log marking an event or operation. The actual information stored includes the operation that will undo the event in case the rewind point is activated. Data recovery uses these rewind points or bookmarks in the Rewind log to reset the current data back to a previous state.
Assured Recovery lets you perform a real test of your disaster recovery server by actually running the application, including modifying data, without impacting your production environment or your previously replicated data.
Using Assured Recovery, you can perform transparent, non-disruptive testing of a replicated data snapshot to start up application services and perform all operations necessary to verify the integrity of the data on the Replica server. Assured Recovery provides this functionality without ever leaving your production systems vulnerable during testing, without disrupting production application availability in any way, and without having to resynchronize the data after testing is complete.
The Suspend mode temporarily stops delivering changes to the suspended Replica server. Changes will continue to be recorded in a spool until replication is resumed so that resynchronization is not required. After replication is resumed, the accumulated changes are then transferred and applied without any need to perform a full resynchronization data.
When you back up a scenario with Assured Recovery configured, the backup is application-consistent and requires no application recovery after restore. However, when you back up a scenario with the Suspend mode enabled (no Assured Recovery configured), the backup may require application recovery after restore, depending upon the state of the application at the time of the backup.
Synchronization is the process of bringing the data on the Replica server in sync with the data on the Master server. To properly synchronize the Master server and the Replica server, their two file structures are compared to determine what content (files and folders) on the Master server is missing or is different from that on the Replica server. The levels of synchronization are as follows:
Replication is a process that maintains identical copies of files and databases by real-time capture of byte-level changes in files on the Master server. These captured changes are asynchronously transmitted to the Replica servers. Because replication continuously updates the source data to another machine, a Replica server always contains the same data as in the Master server. To avoid attempting to restore files that are in use, the application needs to be not running (offline).
Defines the granular level of detail for an Arcserve Replication scenario for backup and restore purposes. The level of granularity for an entity depends on the type of scenario.
For a File Server scenario, an entity represents all of the files and directories belonging to the same volume on a Master server.
For example, on a Master server the contents of the C: drive would constitute one entity, while the contents of the D: drive would be a separate entity.
For a SQL Server scenario, an entity represents a SQL database.
For example, on a Master server the contents of the Company A Employees database would constitute one entity, while the contents of the Company B Employees database would be a separate entity.
For an MS Exchange scenario, an entity represents an MS Exchange Storage Group. The entity name is the MS Exchange Storage Group name.
|
Copyright © 2016
|
|