Global Mirror consistency groups

Global Mirror can control the formation of consistency groups for data consistency.

Global Mirror is based on a combination of Global Copy and FlashCopy functions. To support consistent data across storage units, Global Mirror uses a function called sessions for forming consistent copies. A consistency group is a collection of volumes across multiple storage units that are managed together when creating consistent copies of data. The formation of these consistency groups is coordinated by the master storage unit, which sends commands over remote mirror and copy links to its subordinate storage units.

With Global Mirror functions, consistency groups can be formed many times per hour, as opposed to once or twice per day. By combining many volumes into a session, which is managed by the Global Mirror master storage unit, multiple volumes can be copied to the recovery site simultaneously while maintaining point-in-time consistency across those volumes.

The following properties control how often consistency groups are formed. You can modify these values using the Web-based interface called the IBM^® System Storage™ DS Storage Manager. Access the Global Mirror page, and then Define Properties from the summary of actions menu.

Consistency group interval time: Indicates how long (in seconds) that the storage unit waits between the formation of consistency groups. If you set this value to zero (requires sufficient bandwidth), the consistency group forms continuously, which means that a consistency group starts forming as soon as the previous consistency group has been formed.
Maximum coordination interval: Indicates the maximum time (in milliseconds) that the master storage unit communicates with its subordinate storage units to form a consistent data point. The default is 50 milliseconds.
Because host write I/O is delayed while the consistent point is being set, performance might be affected by allowing too much time for this interval. If the time set for the maximum coordination interval expires before formation of the consistent data point is complete, the consistency group fails.
Maximum time writes are inhibited to the remote site: Indicates the maximum time (in seconds) that write operations are not allowed at the recovery site before the storage unit stops forming a current consistency group. If the drain time is maintained for an extended period of time, the number of write operations that are required to transfer the data to the recovery site can become large enough to increase the time to form a consistency group.
Note: As distance increases, there is an increasing lag in the time it takes for the data to be written to the recovery site. This lag is referred to as a drain time.

Loss of data can also be increased in the event of a disaster. If it is set to zero, it defaults to either four minutes or twice the consistency group interval value, depending on which value is larger. The first consistency group is formed regardless of the consistency group drain time. For the remainder of the consistency groups, if the specified time expires before the data is drained to the recovery site, consistency group formation stops. After the consistency group formation has stopped five consecutive times, the timer is disabled and the next consistency group is formed regardless of the required time.